Chapter 40. Status Asthmaticus

• Severe asthma exacerbation is defined by several, but not necessarily all, of the following features: dyspnea at rest, upright positioning, inability to speak in phrases or sentences, respiratory rate >30 breaths per minute (bpm), use of accessory muscles of respiration, pulse >120 beats/min, pulsus paradoxus >25 mm Hg, peak expiratory flow rate <50% predicted or personal best, hypoxemia, and eucapnia or hypercapnia.
• Altered mental status, paradoxical respirations, bradycardia, a quiet chest, and absence of pulsus paradoxus from respiratory muscle fatigue identify imminent respiratory arrest.
• Airway wall inflammation, bronchospasm, and intraluminal mucus cause progressive airflow obstruction. Fewer patients develop sudden-onset asthma from a more pure form of bronchospasm.
• Severe exacerbations and death can occur in patients with mild asthma.
• Airflow obstruction causes ventilation-perfusion inequality, lung hyperinflation, and increased work of breathing.
• Oxygen, β agonists, and systemic corticosteroids are first-line treatments. Agents of debated efficacy include ipratropium bromide, magnesium sulfate, leukotriene modifiers, theophylline, inhaled steroids, and heliox.
• Noninvasive ventilation is potentially useful in hypercapneic patients not requiring intubation.
• Postintubation hyperinflation decreases preload to the right ventricle and may cause barotrauma, hypotension, or cardiac arrest. A ventilator strategy that lowers lung volume decreases complications.
• Treating airflow obstruction and prolonging expiratory time decreases lung hyperinflation. Expiratory time is prolonged on the ventilator by lowering minute ventilation (even at the cost of hypercapnia) and increasing inspiratory flow rate.
• Deep sedation allows for safe and effective mechanical ventilation in most intubated patients. Paralysis increases the risk of myopathy.
• Patient education, environmental control, and pharmacotherapy are crucial to prevent exacerbations and death.

Asthma is characterized by wheezing, dyspnea, cough, hyperreactive airways, and reversible airflow obstruction.1–3 In the vast majority of cases, the disease is managed uneventfully in the ambulatory setting. However, severe attacks and death can occur regardless of severity classification, and in some cases there is little warning before a terminal event.4 Deaths often occur outside the hospital; patients who have not arrested prior to emergency department (ED) arrival generally have a good prognosis.5 Underestimation of severity, a gap in communication between health care provider and patient, airway inflammation (even in mild disease), and failure to use inhaled corticosteroids all contribute to asthma morbidity and mortality.6–10

Severe asthma exacerbation is defined by several, but not necessarily all, of the following features: dyspnea at rest, upright positioning, inability to speak in phrases or sentences, respiratory rate >30 bpm, use of accessory muscles of respiration, pulse >120/min, pulsus paradoxus >25 mm Hg, peak expiratory flow rate <50% predicted or personal best, hypoxemia, and eucapnia or hypercapnia.1 Altered mental status, paradoxical respirations, bradycardia, a quiet chest, and absence of pulsus paradoxus from respiratory muscle fatigue identify imminent respiratory arrest.1

The objectives of this chapter are to review the pathophysiology of severe asthma exacerbations, discuss clinical assessment and differential diagnosis, assess pharmacologic treatment, and provide a ventilatory strategy for intubated patients. It is perhaps in this last area of acute severe asthma management that intensivists have made their greatest strides over the last two decades, with indirect evidence suggesting that current strategies decrease morbidity and mortality. In the final analysis, however, the greatest gains will come from optimal outpatient management and prevention of attacks.

Exacerbations often evolve over hours to days before patients seek medical help.11,12 There is thus a window of opportunity to treat inflammation through initiation or escalation of corticosteroids. Indeed, the “best treatment of status asthmaticus is to treat it three days before it occurs [with corticosteroids].”13 Instead, too many patients rely on increasing doses of inhaled β agonists, eventually to no avail.

One consequence of airway inflammation is plugging of large and small airways with tenacious mucus. This can be quite striking at postmortem analysis.14 Mucus plugs consist of sloughed epithelial cells, eosinophils, fibrin, and other serum components that leak readily through the denuded airway epithelium. In a smaller subset of patients, an unexpected exacerbation, termed sudden-onset asthma, results from a more pure form of smooth muscle–mediated bronchospasm. Sudden-onset asthma can be lethal, but it can also improve rapidly with bronchodilators.15,16 Compared to attacks of slower tempo, there are more submucosal neutrophils and fewer airway secretions.17,18 Triggers of sudden attacks include use of nonsteroidal anti-inflammatory agents and β-blockers in susceptible patients, allergen or irritant exposure, exercise, stress, sulfites, and inhalation of crack cocaine or heroin.19–22 Respiratory tract infection is not a usual trigger; commonly no cause is found.23

Abnormalities of Gas Exchange

Airway obstruction causes ventilation-to-perfusion (V/Q) mismatch. Intrapulmonary shunting is trivial, so modest enrichment of oxygen (1 to 3 L/min by nasal cannula) generally corrects hypoxemia.24 Refractory hypoxemia is rare and suggests additional pathology: pneumonia, aspiration, acute lobar atelectasis, or barotrauma. There is a rough correlation between airflow obstruction as measured by the forced expiratory volume in 1 second (FEV1) or peak expiratory flow rate (PEFR) and hypoxemia.25,26 However, there is no cutoff value that accurately predicts hypoxemia. In recovering patients, multiple inert gas elimination technique (MIGET) analysis demonstrates that airflow rates commonly improve before partial pressure of arterial oxygen (PaO2) and V/Q inequality.27 One explanation for this is that spirometry tracks large-airway function and indices of gas exchange reflect the function of smaller peripheral airways.27,28

Supplemental oxygen improves oxygen delivery to respiratory muscles, reverses hypoxic pulmonary vasoconstriction, and bronchodilates airways. It further protects from β-agonist–induced hypoxemia resulting from pulmonary vasodilation and increased blood flow to low V/Q units.29,30

Increases in dead space are expected when regions of hyperinflated lung limit blood flow to create West's zone 1 conditions (where alveolar pressure exceeds pulmonary capillary pressure).27 However, MIGET analysis demonstrates only small areas of high V/Q and slightly increased dead space in acute asthma.24,30

Respiratory alkalosis is a feature of mild acute asthma. If present long enough, there is compensatory renal bicarbonate wasting that may subsequently manifest as a normal anion-gap metabolic acidosis (posthypocapneic metabolic acidosis). As the severity of airflow obstruction increases, the partial pressure of arterial carbon dioxide (PaCO2) generally increases due to inadequate alveolar ventilation (reflecting a decrease in minute ventilation as the patient nears respiratory arrest) and possible elevated CO2 production from increased work of breathing. Hypercapnia usually does not occur unless the FEV1 is less than 25% of predicted.25 Importantly the absence of hypercapnia does not preclude a severe attack or potential for respiratory arrest.26

Lung Mechanical Abnormalities

Incomplete emptying of alveolar gas and positive end-expiratory alveolar pressure are hallmarks of expiratory airflow obstruction. In ventilated patients, end-expiratory alveolar pressure is not reflected at the airway opening if the expiratory port of the ventilator is open (which allows airway-opening pressure to approach atmospheric pressure or the level of ventilator-applied positive end-expiratory pressure [PEEP]). If the expiratory port is closed at end-expiration, central airway pressure generally equilibrates with alveolar pressure, permitting measurement of auto- or intrinsic PEEP (PEEPi). PEEPi reflects gas trapping only in relaxed patients, since expiratory muscle contraction elevates end-expired pressure. Low levels of PEEPi may underestimate the degree of lung hyperinflation. Leatherman and Ravenscraft31 measured low levels of PEEPi in four ventilated asthmatics with significant lung hyperinflation, suggesting the presence of noncommunicating airspaces.

The pressure-volume relationship of the lung parenchyma predicts that hyperinflation correlates with decreased static compliance. However, lung compliance may be normal despite hyperinflation, suggesting a stretch-relaxation response in parenchymal tissue.32 This state is not favorable to expiratory flow, but may protect against complications of lung hyperinflation.

Circulatory Effects of Severe Airway Obstruction

Circulatory abnormalities reflect a state of cardiac tamponade resulting from dynamic hyperinflation (DHI) and pleural pressure changes associated with breathing against obstructed airways. During expiration, high intrathoracic pressures decrease right-sided preload. Vigorous inspiration augments right ventricular filling and shifts the intraventricular septum leftward to cause a conformational change in the left ventricle (LV), diastolic dysfunction, and incomplete LV filling. Additionally, large negative pleural pressures directly impair LV emptying.33,34 Rarely diastolic dysfunction and increased LV afterload cause pulmonary edema. Finally, lung hyperinflation increases RV afterload and may cause transient pulmonary hypertension.35 The net effect of these cyclical events is to accentuate the normal inspiratory reduction in stroke volume, a phenomenon termed pulsus paradoxus (PP). Pulsus paradoxus is a valuable marker of asthma severity;36 however, the absence of a widened PP does not ensure a mild attack.37 The PP falls in improving patients, but also in the fatiguing asthmatic who is unable to generate large changes in pleural pressure.

Progression to Ventilatory Failure

Several pathophysiologic mechanisms appear to be responsible for ventilatory failure in acute asthma. PEEPi is a threshold pressure that must be overcome before inspiratory flow occurs, increasing inspiratory work of breathing. Airway resistance is increased, and if lung compliance is low, greater transpulmonary pressure is required. Work of breathing can overwhelm the fatiguing respiratory muscles to cause ventilatory failure, even in young and otherwise healthy individuals.

These mechanical loads are imposed on a diaphragm that is placed in a disadvantageous position by hyperinflation and at a time when circulatory abnormalities may result in hypoperfusion. Respiratory acidosis further decreases diaphragm force generation.38 Lactic acidosis is a marker of severity rather than a cause of diaphragm dysfunction per se.38–40

Multifactorial analysis including the history, physical examination, measures of airflow obstruction, response to therapy, arterial blood gases, and chest radiography is important to assess disease severity and risk for deterioration. Such analysis is necessary because no single clinical measurement predicts outcome reliably.41

Medical History

Characteristics of prior exacerbations that predict a fatal or near fatal episode include intubation, hypercapnia, barotrauma, hospitalization despite corticosteroids, psychiatric illness, and medical noncompliance.42,43 Prior intubation is the greatest single predictor of subsequent asthma death.44 Substance abuse, alcohol ingestion, and excessive, long-term use of β agonists are also associated with increased risk of death.45,46 Pharmacogenetic studies have suggested an association between polymorphisms of β adrenoreceptors, the severity of asthma, and its response to therapy.47 Survivors of near-fatal asthma may also have diminished ventilatory drive in the face of hypoxemic or mechanical stimuli (i.e., poor perceivers of airflow obstruction).48 Other concerning features include symptoms of long duration (which suggests a substantial component of airway wall inflammation and slow recovery), late arrival for care, fatigue, altered mental status, and sleep deprivation. Deterioration despite optimal treatment, including the concurrent use of oral steroids, identifies patients who are unlikely to improve quickly.

“All that wheezes is not asthma” is an appropriate clinical aphorism to consider. In most cases, the history and examination identify conditions that are confused for or complicate acute asthma. The absence of a history of asthma should alert the physician to other diagnoses (although asthma can first occur at any age). A history of smoking suggests chronic obstructive pulmonary disease (COPD), which may be associated with fixed airflow obstruction, pulmonary hypertension, and chronic respiratory acidosis. Cardiac asthma refers to the airway hyperreactivity that occurs in congestive heart failure.49 Heart failure is generally discernible by examination, but the distinction between LV dysfunction and airway obstruction can be difficult. As discussed above, severe airflow obstruction is a rare cause of pulmonary edema, and bronchodilators may improve airflow obstruction in LV failure.50 Foreign body aspiration must be considered in children and in adults with an abnormal swallow, a history of altered mental status, or recent dental work. Upper airway obstruction from granulation tissue, tumor, laryngeal edema, or vocal cord dysfunction may be confused for acute asthma. In classic extrathoracic obstruction, there is flattening of the inspiratory portion of the flow-volume loop and poor response to asthma medication. Fiberoptic laryngoscopy can confirm laryngeal obstruction. For patients with tracheal stenosis (e.g., from prior intubation), CT imaging and fiberoptic bronchoscopy establishes the diagnosis. Important clues to focal obstruction include localized wheeze, and rarely, asymmetric hyperinflation on the chest radiograph.51 Pneumonia complicating asthma is unusual but should be considered when there is fever, purulent sputum, localizing signs, and refractory hypoxemia. Purulent sputum in asthma contains an abundance of eosinophils and not polymorphonuclear leukocytes. In large series of patients with pulmonary embolus, wheezing was not a reported sign. However, this association has been described anecdotally.52 If dyspnea is out of proportion to measures of airflow obstruction, this diagnosis should be considered.

Physical Examination

The general appearance of the patient (i.e., posture, pattern of speech, positioning, and level of alertness) allows for quick assessment of status, response to therapy, and need for intubation. Adults who assume the upright position have a higher heart rate (HR), respiratory rate (RR), and PP and a significantly lower PaO2 and PEFR than patients who are able to lie supine.53 Diaphoresis is associated with an even lower PEFR. Accessory muscle use and PP indicate severe airflow obstruction; the absence of either finding does not rule out severe obstruction.37

Examination of the head and neck should focus on identifying barotrauma and upper airway obstruction. Tracheal deviation, asymmetric breath sounds, a “mediastinal crunch,” and subcutaneous emphysema suggest pneumomediastinum or pneumothorax. Rarely, atelectasis from a central mucus plug causes tracheal deviation. The mouth and neck should be inspected for mass lesions or signs of previous surgery. The lip and tongue should be inspected for angioedema.

Wheezing correlates poorly with the degree of airflow limitation.54 A silent chest suggests insufficient airflow for wheezes to occur, and in general, wheezes that emerge in a previously silent chest signal clinical improvement. Localized wheezing or crackles should prompt consideration of atelectasis, pneumonia, pneumothorax, endobronchial lesions, or foreign body.

Asthmatics are generally tachycardic on presentation.55 Bradycardia is an ominous sign of impending arrest. The usual rhythm is sinus tachycardia. Supraventricular and ventricular arrhythmias are not uncommon, particularly in older patients.56

Clinical signs of right-sided and left-sided heart failure suggest primary cardiac disease. Yet, acute asthma alone can cause examination and electrocardiographic findings of right-sided heart strain and pulmonary edema which resolve within hours of response to therapy.55,57 Jugular venous distention also occurs when DHI or tension pneumothorax limit venous return to the right heart. Myocardial ischemia should be considered in older patients with coronary artery disease. Asthma may cause myocardial oxygen supply/demand imbalance if large drops in intrathoracic pressure increase LV afterload and decrease coronary blood flow.58 β Agonists, theophylline, and hypoxemia may further disrupt this balance.

Measurement of Airflow Obstruction

The degree of airflow obstruction is determined by measuring PEFR or FEV1. A PEFR or FEV1 <50% of predicted or the patient's personal best characterizes severe exacerbation.1 Objective measures are necessary because physician estimates are often wrong.59 In most cases it is easier to measure PEFR than FEV1, but neither is easy to obtain in critical patients. In these patients it is wise to defer measurements, since they may worsen bronchospasm60 and rarely precipitate respiratory arrest.61 In most cases, however, peak flow determinations are safe.

Measurement of the change in PEFR or FEV1 is a good way to assess and follow patients. Several studies have demonstrated that failure of initial therapy to improve expiratory flow after 30 minutes predicts a refractory course and need for hospitalization or continued treatment in an ED.41,62–64 Changes in PEFR before 30 minutes of treatment have elapsed do not predict clinical outcome.65

Arterial Blood Gases

When FEV1 is less than 50% predicted or of the patient's personal best, an arterial blood gas should be considered. In early (mild) acute asthma, mild hypoxemia and respiratory alkalosis are common. Hypercapnia indicates severe disease and impending respiratory failure, but not necessarily the need for intubation.66 Importantly, hypercapnia is not always present in cases of severe obstruction and impending respiratory arrest.67

Metabolic acidosis with a normal anion gap occurs when there is bicarbonate wasting in response to respiratory alkalosis. An elevated anion gap suggests excess serum lactate, possibly secondary to increased work of breathing, tissue hypoxia, intracellular alkalosis, or decreased lactate clearance by the liver. Lactic acidosis correlates with the severity of airflow obstruction, is more common in men, and occurs more frequently when β agonists are administered parenterally.68,39

Serial blood gases are usually not necessary to determine clinical course. Physical examinations and peak flows allow for valid judgements in most cases. Patients who deteriorate on clinical grounds with impending respiratory failure should be intubated regardless of PaCO2. Conversely, improving patients should not be intubated despite hypercapnia. Serial blood gases are helpful in intubated patients to guide ventilator management.

Radiographic Studies

Chest radiography plays little role in the assessment or management of patients with acute asthma. Even in hospitalized patients, radiographic findings influence treatment in 1% to 5% of cases.69–72 In one study72 that reported major radiographic abnormalities in 34% of cases (which the authors felt impacted management), the majority of findings were classified as focal parenchymal opacities or increased interstitial markings, common indicators of atelectasis in asthma. Chest radiography is indicated when there are localizing signs, barotrauma is suspected, or it is not clear that asthma is the cause of respiratory distress. In mechanically ventilated patients, chest radiography further identifies endotracheal tube position.

Admission Criteria

Patients demonstrating a good response to initial therapy in the ED may be discharged home with close follow-up. There should be significant improvement in breathlessness, improved air movement on physical examination, and a FEV1 or PEFR ≥70% of predicted or personal best.1 Observation for 60 minutes after the last β-agonist dose helps ensure stability prior to discharge. Written medication instructions and an action plan to be followed in the event of worsening symptoms should be provided. In general, patients should be discharged on oral corticosteroids. We advise initiation and/or continuation of inhaled corticosteroids by proper technique in the ED. Follow-up appointments are essential.

Patients with severe airflow obstruction demonstrating a poor response to therapy (e.g., <10% increase in PEFR) or patients who deteriorate despite therapy should be admitted to an ICU. Other indications for ICU admission include respiratory arrest, altered mental status, arrhythmias, myocardial ischemia or infarction, and need for frequent inhaler treatments.

An incomplete response to treatment is defined as the persistence of wheezing or dyspnea and a PEFR or FEV1 between 50% and 70% of predicted.1 Patients in this group require ongoing treatment either in the ED or medical ward. In general, we treat for up to 6 hours in the ED before deciding on disposition. This length of time allows for assessment of the initial response to steroids, and for those who are discharged home decreases the risk of relapse and return to the ED.37 If the patient has demonstrated a good response to treatment over that period of time and close follow-up can be established, discharge home is appropriate. Discharged patients not optimally treated prior to the ED have better outcomes if they receive a course of oral steroids.73 Patients who continue to have an incomplete response should be admitted to a medical ward. Admission is also recommended when there is a harmful home environment or when noncompliance favors directly observed therapy.

Pharmacotherapy

Beta Agonists

Asthmatic smooth muscle has increased shortening capacity and increased velocity of shortening74 that mandate treatment in the setting of an acute attack. Use of a β agonist by inhalation is preferred and should begin immediately regardless of prior use (Table 40-1).75 Large and frequent doses are needed in acute asthma because the dose-response curve and duration of action are adversely affected by airway narrowing. Fortunately, high doses are generally well tolerated. In one study, albuterol delivered by metered dose inhaler (MDI) and spacer to a total dose of 1600 μg over 90 minutes did not increase cardiovascular morbidity in well-oxygenated patients with acute asthma.76

Albuterol is preferred over metaproterenol because of its greater β2 selectivity and longer duration of action.77,78 Some clinicians prefer metaproterenol or isoetharine for initial therapy, despite the tendency of these drugs to increase side effects, because of their faster onset of action.79  Levalbuterol is the R-isomer of racemic albuterol. Preclinical data demonstrate that the S-isomer has undesirable proinflammatory effects.80 The clinical relevance of these effects (particularly in those receiving high doses of the racemate) is under investigation. Clinical data do show that 1.25 mg levalbuterol increases FEV1 to a greater extent than 2.5 mg of the racemate in stable patients with FEV1 ≤60% of predicted.81 In acute asthma, an open-label study of 91 patients with an FEV1 of 20% to 55% of predicted, stratified into cohorts receiving 0.63 to 5 mg levalbuterol or 2.5 to 5 mg albuterol (three treatments in 1 hour), demonstrated faster onset of action and a greater degree of bronchodilation with levalbuterol.82 Results from a larger multicenter study are expected in the near future.

Long-acting β agonists are not indicated in the initial treatment of acute asthma, but may be considered as add-on therapy in hospitalized patients. The addition of salmeterol to albuterol in hospitalized asthmatics is safe and improves FEV1 after 48 hours compared with placebo.83

Subcutaneous administration is not indicated unless the patient is unable to carry out inhaled therapy (such as those with an altered mental status or cardiopulmonary arrest), and is associated with greater toxicity.84–86 Subcutaneous epinephrine, however, benefits some patients not responding to several hours of an inhaled β agonist.87 Known cardiac disease and age >40 years are relative contraindications to parenteral therapy,88 although older patients without recent myocardial ischemia or infarction tolerate subcutaneous epinephrine reasonably well. Intravenous infusion of β agonists is not recommended. Inhaled drug achieves greater airflow rates with lesser toxicity compared to intravenous drug.89–92 Combinations of inhaled and intravenous treatment have not been adequately studied.

Inhaled β agonists are delivered equally well by MDI with spacer or hand-held nebulizer.93 Even in severe obstruction, four puffs of albuterol (0.36 mg) delivered by MDI with an InspirEase is as effective as 2.5 mg of albuterol by nebulization.94 MDIs with spacers are less expensive and able to achieve faster drug delivery times; hand-held nebulizers require fewer instructions, less supervision, and less coordination.

The recommended initial dose is albuterol 2.5 mg by nebulization; three treatments can be given in the first hour depending on clinical response and side effects. A single high-dose treatment (e.g., 7.5 mg) is no better than multiple treatments (2.5 mg every 20 minutes 3 times), and causes more side effects.95 Depending on the study, there is either no difference in efficacy or toxicity between continuous or intermittently nebulized albuterol, or a slight benefit to continuous nebulization (at the same total dose per hour) in severe obstruction.96–98

There are two dose response patterns to high and cumulative doses of albuterol. Approximately two-thirds of patients respond to inhaled albuterol in a dose-dependent fashion, generally allowing discharge home from the ED.99 In these patients 1.2 to 2.4 mg albuterol delivered by MDI and spacer or 5 to 7.5 mg by nebulizer in the first hour is effective. In the remaining one-third of patients, even in high doses, albuterol has minimal effect, presumably because of airway inflammation and mucus plugging.

Ipratropium Bromide

Data support the addition of ipratropium bromide to albuterol in acute asthma.100–106 Karpel and colleagues studied 384 patients randomized to nebulized albuterol 2.5 mg or albuterol 2.5 mg with ipratropium 0.5 mg at entry and at 45 minutes.107 At 45 minutes, there were more responders in the ipratropium group; however, the median change of FEV1 from baseline did not differ between groups, and by 90 minutes there was no difference between groups in the percentage of responders and median change in FEV1. There was no difference in the number of patients requiring additional ED or hospital treatment. Garrett and colleagues randomized 338 asthmatics to a single dose of nebulized ipratropium bromide 0.5 mg combined with salbutamol 3.0 mg, or salbutamol 3.0 mg alone.108 Mean FEV1 at 45 and 90 minutes was significantly higher with combined therapy. Lin and colleagues demonstrated that combination therapy resulted in greater improvement in PEFR than albuterol alone in 55 adult asthmatics.109 O'Driscoll and colleagues also demonstrated a benefit to combination therapy, particularly in patients with PEFR <140 L/min.110 In children, combination therapy decreases ED treatment time, albuterol dose requirements, and hospitalization rates.106,111,112

To the contrary, Weber and colleagues performed a prospective, randomized, double-blind, placebo-controlled trial of 67 patients receiving combination albuterol and ipratropium bromide or albuterol alone by continuous nebulization for a maximum of 3 hours.113 Primary outcome measures were improvement in PEFR, hospital admission rates, and length of stay in the ED. All outcome measures favored combination therapy, but differences were not statistically significant. Fitzgerald and coworkers reported similar results.114 McFadden and colleagues showed no benefit in PEFR, admission rate, or ED length of stay with combination therapy used for the first hour.115 In children, Ducharme and Davis did not demonstrate benefit from combination therapy in their study of nearly 300 asthmatics with mild to moderate acute asthma.116

The above studies demonstrating modest (or no) benefit to combination therapy generally used small doses of ipratropium bromide. Few trials have studied high and cumulative doses of both ipratropium bromide and albuterol. In one such study, Rodrigo and Rodrigo conducted a double-blind, randomized, prospective trial of albuterol and ipratropium bromide (120 μg of albuterol and 21 μg of ipratropium per puff, combined in one inhaler) vs. albuterol (120 μg per puff) and placebo in 180 patients with acute asthma.117 Four puffs were administered through an MDI with a spacer every 10 minutes for 3 hours. Combination therapy resulted in a 20.5% greater improvement in PEFR and a 48.1% greater improvement in FEV1 compared with albuterol alone. The rate of hospitalization decreased significantly, from 39% in the albuterol with placebo group to 20% with combination therapy. Subgroup analysis showed that patients most likely to benefit from high doses of ipratropium bromide were those with FEV1 <30% of predicted and symptoms for >24 hours prior to ED presentation.

Overall, the data suggest an advantage in maximal bronchodilation response when high doses of ipratropium bromide and albuterol are combined in the emergency treatment of asthma. Combination therapy is recommended in any patient who is extremely ill on first presentation or not responding quickly (e.g., within 30 minutes) to albuterol alone.

Corticosteroids

Systemic corticosteroids should be administered quickly to treat the inflammatory component of acute asthma. The data are mixed regarding benefit in the first few hours. McFadden's group demonstrated no differences in physiologic or clinical variables in the first 6 hours in 38 patients receiving hydrocortisone.118 Rodrigo and Rodrigo similarly showed that early administration of steroids did not improve spirometry in the first 6 hours.119 However, Littenberg and Gluck demonstrated that methylprednisolone 125 mg IV on arrival decreased admission rates compared to placebo,120 and Lin and colleagues demonstrated improved peak flows after 1 and 2 hours of solumedrol.121 A systematic review of 12 studies for the Cochrane Review demonstrated that corticosteroids within 1 hour of arrival in the ED reduced admissions.122 Use of early systemic steroids also reduces the number of relapses in the first 7 to 10 days and the risk of death.123–126 In hospitalized patients, systemic steroids speed the rate of improvement.127 Oral and intravenous routes are equally effective,128 but oral steroids should be avoided if there is concern regarding intubation.

There is no clear dose-response relationship to steroids in acute asthma.129,130 In one meta-analysis by Manser and colleagues, there was no difference in clinical outcomes between low-dose corticosteroids (≤80 mg/d solumedrol or ≤400 mg/d hydrocortisone) and higher doses in the initial management of hospitalized asthmatics.131 Haskell and colleagues demonstrated that 125 mg IV methylprednisolone every 6 hours resulted in faster improvement compared to 40 mg every 6 hours, but there was no difference in peak improvement.132 Both doses were superior to 15 mg every 6 hours in terms of rate and absolute response. Emerman and Cydulka compared 500 mg and 100 mg of methylprednisolone and found no benefit to the higher dose.133

The Expert Panel from the National Institutes of Health (NIH) recommends 120 to 180 mg per day of prednisone or methylprednisolone in three or four divided doses for 48 hours, then 60 to 80 mg/d until PEFR reaches 70% predicted or personal best.1 Prednisone is tapered at variable rates depending on a number of factors, including PEFR, the duration of high-dose therapy required to treat the acute exacerbation, and whether oral steroids had been used for maintenance therapy. Automatic tapering schedules are not recommended because patients may taper prematurely.

Recent trials have demonstrated efficacy of inhaled corticosteroids in acute asthma. In children discharged from the ED, short-term use of budesonide, starting at a high-dose and tapered over 1 week, was as effective as a tapering course of oral prednisolone.134 Rodrigo and Rodrigo conducted a randomized, double-blind trial of the addition of flunisolide 1 mg or placebo to 400 μg salbutamol every 10 minutes for 3 hours in 94 ED subjects not receiving systemic steroids.135 They found that PEFR and FEV1 were approximately 20% higher in the flunisolide group, beginning at 90 minutes. McFadden suggests this benefit stems from steroid-induced vasoconstriction, decreasing airway wall edema, vascular congestion, and plasma exudation, and not anti-inflammatory effects.136 Recently Rodrigo and Rodrigo demonstrated therapeutic benefit from triple drug therapy (flunisolide, albuterol, and ipratropium bromide) in high doses in patients not receiving systemic steroids.137

To the contrary, Guttman and colleagues found no benefit from the addition of beclomethasone 7 mg every 8 hours by MDI with spacer to nebulized salbutamol and systemic corticosteroid therapy.138 Further study demonstrated that beclomethasone (5 mg delivered by MDI) during the initial 4 hours of ER treatment did not confer added benefit to albuterol in adults with mild to moderately severe asthma.139

Our view is that there is little benefit to the addition of inhaled steroids to high-dose β agonists and systemic corticosteroids in the management of acute asthma. Still, consideration should be given to the use of high-dose inhaled corticosteroids in refractory patients.

Aminophylline

There is no benefit to adding aminophylline to inhaled β agonists in the initial treatment of acute asthma.140 In a meta-analysis by Parameswaran and colleagues there was a trend toward higher PEFR at 12 and 24 hours, but at the cost of arrhythmias and vomiting.141 Others have reported a delayed benefit.142 In another recent meta-analysis in children, aminophylline was shown to improve FEV1 by 6 to 8 hours and provide sustained benefit for 24 hours.143 Nonbronchodilating properties of aminophylline may be useful in refractory cases. Anti-inflammatory effects and enhanced diaphragm function may explain one report that ED administration of aminophylline decreased hospitalizations, even when airflow rates were no different than placebo.144

We rarely use aminophylline in refractory patients, but generally continue its use (after confirming a nontoxic serum concentration) in patients already taking theophylline. This approach is safe if attention is paid to serum drug levels and to factors that increase levels, such as congestive heart failure, ciprofloxacin, macrolide antibiotics, and cimetidine, and if the drug is discontinued in the presence of side effects.

Magnesium Sulfate

Three prospective trials failed to confirm a benefit to administering magnesium sulfate (MgSO4) to asthmatics in the ED.145–147 In 135 asthmatics randomized to 2 g MgSO4IV or placebo after 30 minutes and followed for 4 hours, admission rates and FEV1 were no different between magnesium-treated patients and controls.147 However, subgroup analysis revealed MgSO4 decreased admission rates and improved FEV1 in subjects with FEV1 <25% of predicted. Subsequently, a placebo-controlled, double-blind, randomized trial in 248 patients with FEV1 ≤30% showed a small but statistically significant increase in FEV1 after 240 minutes in the magnesium group, but no difference in hospitalization rates.148 Meta-analysis of 9 trials and 665 patients demonstrated nonsignificant improvements in peak flows and admission rates for all patients, but in those with severe exacerbations, MgSO4 increased FEV1 by nearly 10% compared to placebo, and decreased admissions.149 Additional evidence supporting benefit in severe disease comes from an uncontrolled study of five intubated asthmatics given magnesium.150 In this study, there was a fall in peak airway pressure (43 cm H2O to 32 cm H2O) after high doses of MgSO4 (10 to 20 g) were administered over 1 hour. Other investigators have suggested that gender may play a role in magnesium responsiveness, since estrogen augments the bronchodilator effect of magnesium.151 Routine use of MgSO4 is not justified, but MgSO4 is safe, inexpensive, and may be beneficial in severe exacerbations.

Magnesium sulfate can also be administered by inhalation. Nannini and colleagues studied the effects of MgSO4 (225 mg) vs. saline as the vehicle for nebulized albuterol in a randomized, double-blind fashion.152 At 20 minutes, patients treated with MgSO4 and albuterol had a greater PEFR compared to the saline-albuterol group (134 ± 70 L/m vs. 86 ± 64 L/m). Hughes and colleagues have published similar data.153

Leukotriene Modifiers

Cysteinyl leukotrienes are elevated in asthmatic sputum compared to controls and higher in subjects studied within 48 hours of exacerbation.154 Preliminary data demonstrating benefit to a leukotriene receptor antagonist came from a double-blind, randomized trial of two doses (20 mg and 160 mg) of zafirlukast orally versus placebo in 641 asthmatics after 30 minutes of standard treatment.155 Zafirlukast 160 mg decreased admission rates, relapses, and treatment failures. In another double-blind, placebo-controlled study of 20 patients not receiving systemic steroids in an ED, oral montelukast 10 mg resulted in a trend toward a shorter duration of stay and higher peak flows, and fewer patients requiring aminophylline or steroids.156 In the most compelling trial to date, Camargo and colleagues randomized 201 acute asthmatics to standard therapy plus montelukast 7 mg or 14 mg IV or placebo.157 Montelukast improved FEV1 over the first 20 minutes (14.8% vs. 3.6% with placebo). Benefits were seen within 10 minutes and lasted for 2 hours; both treatment doses were equivalent. Montelukast also tended to result in less β-agonist use and fewer treatment failures.

Heliox

Heliox is a mixture of 20% oxygen and 80% helium (30%:70% and 40%:60% mixtures are also available). As the percentage of helium decreases, so does the benefit of breathing this gas blend. Concentrations of helium less than 60% are ineffective, precluding its use in significant hypoxemia. Heliox is slightly more viscous than air, but significantly less dense, resulting in a more than threefold increase in kinematic viscosity (the ratio of gas viscosity to gas density) compared to air. Theoretically, this property decreases the driving pressure required for gas flow by two mechanisms. First, for any level of turbulent flow, breathing low-density gas decreases the pressure gradient required for flow. Second, heliox decreases the Reynolds number, favoring conversion of turbulent flow to laminar flow.158 Heliox does not treat bronchospasm or airway wall inflammation.

Heliox promptly improves dyspnea, work of breathing, and arterial blood gases in upper airway obstruction.159 Benefits have also been reported in acute asthma. In adults treated in an ED, an 80:20 mix delivered by tight-fitting face mask increased PEFR and decreased PP, suggesting improved airway resistance and work of breathing.160 Similar results have been published in children.161 Other studies have failed to demonstrate benefit.162–164 In a recent meta-analysis by Rodrigo and colleagues of four randomized trials and 288 patients, the authors concluded that the evidence does not support the use of heliox in nonintubated asthmatics in the ED.165 However, methodologic differences between studies limit the ability to draw conclusions.

If heliox is effective, it may give time for concurrent therapies to work, and thereby avert the need for intubation in some cases. Of theoretical concern is the potential for heliox to mask worsening airflow obstruction, so there is less time (and no margin for error) to control the airway when intubation is required.

Whether heliox augments the bronchodilator effect of inhaled β agonists compared to delivery in air (presumably due to low-density gas facilitating albuterol deposition) is unclear. Data are available demonstrating benefit to heliox as a driving gas,166 but there are also data to the contrary.167

Antibiotics

Because viruses trigger most infectious exacerbations of asthma and bacterial pneumonia is rare, there is likely no role for antibiotics in treating acute asthma. Antibiotics are frequently prescribed for an increase in sputum volume and purulence. However, purulence may reflect an abundance of eosinophils, not polymorphonuclear leukocytes. The importance of Mycoplasma pneumoniae in acute asthma is unknown. A recent study by Lieberman using paired serology demonstrated evidence for mycoplasma infection in 18% of patients hospitalized for acute asthma.168 In their 2002 update, the Expert Panel from the NIH did not recommend the use of antibiotics in asthma exacerbation unless there was fever with purulent sputum, evidence for pneumonia, or suspected bacterial sinusitis.169 In a separate review of the literature, Graham and associates recently selected 2 out of 128 possible studies adequate for review, concluding that the role of antibiotics is difficult to assess.170

Noninvasive Positive Pressure Ventilation

Noninvasive positive pressure ventilation (NPPV) by face mask is an option for patients with hypercapneic respiratory failure who do not require intubation. Continuous positive airway pressure (CPAP) helps overcome the adverse effects of PEEPi and decreases the inspiratory work of breathing.171 Bronchial dilation also occurs during CPAP.172 Advantages of NPPV over intubation include decreased need for sedation and paralysis, decreased incidence of nosocomial pneumonia, decreased incidence of otitis and sinusitis, and improved patient comfort.173 Disadvantages include increased risk of aspiration when there is gastric insufflation, skin necrosis, and diminished control of ventilatory status compared with invasive ventilation. Data regarding the efficacy of NPPV in acute asthma are limited. In one study174 of 21 acute asthmatics with a mean PEFR of 144 L/min, nasal CPAP of 5 or 7.5 cm H2O decreased respiratory rate and dyspnea compared to placebo. In another study,175 Meduri and colleagues reported their observational experience with NPPV during 17 episodes of acute severe asthma. NPPV was achieved using a loose-fitting full-face mask with initial settings of 0 cm H2O CPAP and 10 cm H2O pressure support ventilation. CPAP was increased to 3 to 5 cm H2O and pressure support was titrated to achieve an exhaled tidal volume of ≥7 mL/kg and a respiratory rate of <25/min. The average duration of NPPV was 16 hours. NPPV generally improved dyspnea, HR, RR, and blood gases. Two NPPV-treated patients required intubation for worsening PaCO2. There were no NPPV complications.

Intubation

The goals of intubation and mechanical ventilation are to maintain oxygenation and prevent respiratory arrest. Patients who are intubated before they arrest generally do well (if ventilated appropriately); patients who are intubated after respiratory arrest and the development of anoxic brain injury do not. Approximately 40% of deaths in ventilated asthmatics are due to cerebral anoxia resulting from respiratory arrest prior to mechanical ventilation.176

Intubation is indicated for impending respiratory failure and cardiopulmonary arrest. Changes in posture, mental status, speech, accessory muscle use, and RR can indicate progressive ventilatory failure that does not need blood-gas or PEFR confirmation. In the final analysis, the decision to intubate rests on a clinician's estimate of the patient's ability to maintain spontaneous respiration.

Oral intubation is preferred because it allows for placement of an adequately sized endotracheal tube (e.g., 8.0 mm inside diameter [ID] for adult women, 8.0 to 9.0 mm ID for adult men) to facilitate removal of mucus and decrease airflow resistance. Nasal intubation is acceptable in an awake patient anticipated to be difficult to position and intubate (fiberoptic guidance may facilitate intubation in this setting), but is complicated by the need for a smaller endotracheal tube and the high incidence of nasal polyps and sinusitis in asthma.

Postintubation Hypotension

Hypotension has been reported in 25% to 35% of patients following intubation.176 It occurs for several reasons. First, there is loss of vascular tone due to a direct effect of sedation and loss of sympathetic activity. Second, patients may be hypovolemic from high insensible losses and decreased oral fluid intake. Third, overzealous Ambu Bag ventilation results in DHI because adequate time is not provided for exhalation. When this occurs, breath sounds diminish, BP falls, and HR rises. Manual breaths take considerable effort because the hyperinflated lung is stiff (noncompliant). A trial of hypopnea (2 to 3 breaths/min) or apnea in a preoxygenated patient deflates the lung. After 30 to 60 seconds of hypoventilation, intrathoracic pressure falls, allowing for greater preload to the right atrium; BP rises, HR falls, and breaths become easier to deliver, confirming DHI.

A trial of hypopnea or apnea is both diagnostic and therapeutic for DHI. However, hemodynamic improvement during such a trial does not exclude tension pneumothorax. Tension pneumothoraces may have been responsible for >6% of deaths of patients who required mechanical ventilation for severe asthma.176 Careful inspection of the chest x-ray is mandatory because the lungs may not collapse completely in the setting of DHI and widespread mucus plugging. Unilateral pneumothorax conceivably increases ventilation to the contralateral lung, increasing the risk of bilateral pneumothoraces. Given these considerations, chest tubes should not be placed until a trial of hypoventilation has failed or there is radiographic evidence of pneumothorax.

Initial Ventilator Settings and Dynamic Hyperinflation

Expiratory time, tidal volume, and severity of airway obstruction determine the level of DHI (Fig. 40-1). Minute ventilation and inspiratory flow determine expiratory time.177,178 To avoid dangerous levels of DHI, initial minute ventilation should not exceed 115 mL/kg per minute or approximately 8 L/min in a 70-kg patient.179 To this end, we recommend a RR between 12 and 14/min and a tidal volume between 7 and 8 mL/kg. The use of low tidal volumes avoids excessive peak lung inflation, which can occur even with low minute ventilation.

Shortening the inspiratory time by use of a high inspiratory flow rate further prolongs expiratory time. We favor a flow rate of 80 L/min, using a constant flow rate pattern. High inspiratory flows increase peak airway pressure by elevating airway resistive pressure, but peak airway pressure per se does not correlate with morbidity or mortality. Rather it is the state of lung hyperinflation that predicts outcome (see below). High inspiratory flow and high airway pressures may redistribute ventilation to low-resistance lung units, risking barotrauma, but these concerns are based largely on mathematical and mechanical lung models,180,181 and have no supporting clinical data. Another concern in spontaneously breathing patients is that high inspiratory flow rates in the assist-control mode can increase RR and thereby decrease expiratory time.182

There is no consensus as to which ventilator mode should be used in asthmatics. In paralyzed patients, synchronized intermittent mandatory ventilation (SIMV) and assist-controlled ventilation (ACV) are equivalent. In patients triggering the ventilator, SIMV is generally preferred because of the unproven concern that minute ventilation will be higher during ACV, since each triggered breath receives a guaranteed tidal volume.183 However, in spontaneously breathing patients, SIMV may increase work of breathing and machine-patient dyssynchrony if set minute ventilation is low.184 Volume-controlled ventilation (VCV) is recommended over pressure-controlled ventilation (PCV) for several reasons, including staff familiarity with its use. PCV offers the advantage of limiting peak airway pressure to a predetermined set value. However, during PCV, tidal volume is inversely related to PEEPi and minute ventilation is not guaranteed. Peak inspiratory flow rates may be extremely high in PCV (to compensate for decelerating flow) to maintain a short inspiratory time.

Ventilator-applied PEEP is not recommended in sedated and paralyzed patients because it may increase lung volume if used excessively.185 In spontaneously breathing patients, small amounts of ventilator-applied PEEP (e.g., 5 cm H2O) decrease inspiratory work of breathing by decreasing the pressure gradient required to overcome PEEPi.

Assessing Lung Inflation

Determining the level of DHI is central to risk assessment and ventilator adjustment. Numerous methods have been proposed to measure DHI. The volume at end-inspiration, termed VEI, is determined by collecting all expired gas from the end-inspiratory volume to functional reserve capacity (FRC) during 40 to 60 seconds of apnea (Fig. 40-2). Although VEI may underestimate the degree of air trapping if there are very slowly emptying air spaces, VEI greater than 20 mL/kg correlates with barotrauma.179 The utility of this measure is limited by the need for paralysis, and the fact that most clinicians and respiratory therapists are unfamiliar with expiratory gas collection.

Surrogate measures of DHI include the single-breath plateau pressure (Pplat) and PEEPi. Neither is perfect. Pplat is an estimate of average end-inspiratory alveolar pressures that is determined by stopping flow at end-inspiration (Fig. 40-3). PEEPi is the lowest average alveolar pressure achieved during the respiratory cycle. It is obtained by measuring airway-opening pressure during an end-expiratory hold maneuver (Fig. 40-4). In the presence of PEEPi, airway-opening pressure increases to the amount of PEEPi present. Persistence of expiratory gas flow at the beginning of inspiration (which can be detected by auscultation or monitoring of flow tracings) also demonstrates PEEPi.186

Accurate measurement of Pplat and PEEPi requires patient-ventilator synchrony and patient relaxation. Paralysis is generally not required. Unfortunately neither measure has been validated as a predictor of complications. Pplat is affected by the lung and surrounding structures so that variations in DHI occur at the same pressure. For example, an obese patient will have a higher Pplat than a thin patient for the same degree of DHI. Despite these limitations, experience suggests that a Pplat <30 cm H2O is generally safe.

PEEPi may underestimate the severity of DHI (see above). This may occur when airway closure limits communication between the alveolus and airway opening, so that during an end-exhalation hold maneuver, airway-opening pressure does not rise. In most cases, however, PEEPi <15 cm H2O is acceptable.

Ventilator Adjustments

Relying on Pplat as a measure of DHI, we provide the following suggestions for ventilator adjustments. If initial settings result in Pplat >30 cm H2O, RR is decreased until this goal is achieved, even at the cost of hypercapnia. Hypercapnia is generally well tolerated in oxygenated patients, even to PaCO2 values nearing 90 mm Hg, as long as sudden changes do not occur.187,188 Anoxic brain injury and myocardial dysfunction are contraindications to permissive hypercapnia, which may induce cerebral vasodilation, decrease myocardial contractility, and constrict the pulmonary vascular bed.189 Lowering minute ventilation may not cause the expected rise in PCO2 if dead space decreases concurrently.

If hypercapnia results in a blood pH of less than 7.20 (and RR can not be increased because of the Pplat limit), we consider a slow infusion of sodium bicarbonate, although this has not been shown to improve outcome.190 If Pplat is less than 30 cm H2O and pH is less than 7.20, RR can be safely increased for the purpose of lowering PaCO2 and elevating arterial pH until Pplat nears the threshold pressure. Commonly, patients can be ventilated to a pH >7.20 with a Pplat <30 cm H2O.

Sedation and Paralysis

Sedation is indicated to improve comfort, safety, and patient-ventilator synchrony. This is particularly true when hypercapnia serves as a potent stimulus to respiratory drive. Some patients (such as those with sudden-onset asthma) may be ready for extubation within hours. In these patients, propofol is attractive because it can be rapidly titrated to a deep level of sedation, and there is quick reversal of sedation after discontinuation.191  Lorazepam and midazolam are less expensive alternatives.192 Time to awakening after discontinuation of benzodiazepines is longer and less predictable than with propofol.

In order to provide the best combination of amnesia, sedation, analgesia, and suppression of respiratory drive, we recommend the addition of a narcotic by continuous infusion to propofol or a benzodiazepine.193  Morphine and fentanyl are the two most commonly used narcotics. Fentanyl has a quicker onset of action than morphine. Narcotics may release histamine and worsen bronchospasm if they are delivered by bolus administration. Daily interruption of sedatives avoids unwanted drug accumulation.194

Ketamine, an IV anesthetic with sedative, analgesic, and bronchodilating properties, is reserved for use in intubated patients with severe bronchospasm that precludes safe mechanical ventilation.195–197 Ketamine must be used with caution because of its sympathomimetic effects and ability to cause delirium.

When safe and effective mechanical ventilation cannot be achieved by sedation alone, short-term muscle paralysis is indicated. Short- to intermediate-acting agents include atracurium, cis-atracurium, and vecuronium. Pancuronium is longer acting and less expensive, and its vagolytic effects cause tachycardia. Pancuronium and atracurium release histamine, but this is of doubtful clinical significance.198 We prefer cis-atracurium because it is essentially free of cardiovascular effects, does not release histamine, and does not require hepatic and renal function for clearance.

Paralytics may be given intermittently by bolus or continuous IV infusion. If a continuous infusion is used, a nerve stimulator should be used or the drug should be withheld every 4 to 6 hours to avoid drug accumulation. Paralytic agents should be minimized whenever possible because of the risk of postparalytic myopathy.199–201 In one study of 25 ventilated asthmatics, 19 (76%) patients had an increase in serum creatine kinase, and 9 (36%) had clinically detectable myopathy.201 Elevated creatine kinase was associated with prolonged mechanical ventilation whether or not there was clinically detectable myopathy. In a retrospective cohort study of 107 episodes of asthma requiring intubation, the concurrent use of steroids and a paralytic was associated with muscle weakness in 29% of episodes; steroid treatment alone was not associated with weakness.199 Importantly, this study demonstrated that the duration of paralysis correlated with the incidence of myopathy, which is rare when paralytics are used for less than 24 hours. Findings of a separate study confirm this correlation.200 Most patients with postparalytic myopathy recover, but may require weeks of rehabilitation.

Administration of Bronchodilators during Mechanical Ventilation

Many questions remain regarding the administration of inhaled bronchodilators to intubated patients. In one study,202 only 2.9% of a radioactive aerosol delivered by nebulizer was deposited in the lungs of mechanically ventilated patients. Manthous and colleagues compared the efficacy of albuterol delivered by MDI via a simple inspiratory adapter (no spacer) to nebulized albuterol in intubated patients.203 Using the peak-to-pause pressure gradient at a constant inspiratory flow to measure airway resistance, they found no effect (and no side effects) from the administration of 100 puffs (9.0 mg) of albuterol. Albuterol delivered by nebulizer to a total dose of 2.5 mg reduced the inspiratory flow-resistive pressure 18%. Increasing the nebulized dose to a total of 7.5 mg reduced airway resistance further in 8 of 10 patients, but caused side effects in half of the patients. Thus if MDIs are used during mechanical ventilation, use of a spacer on the inspiratory limb of the ventilator improves drug delivery.204

Regardless of whether an MDI with spacer or nebulizer is used, higher drug dosages are required and the dosage should be titrated to achieve a fall in the peak-to-pause airway pressure gradient. Nebulizers should be placed close to the ventilator, and in-line humidifiers stopped during treatments. Inspiratory flow should be reduced to approximately 40 L/min during treatments to minimize turbulence, although this strategy may worsen DHI and must be time-limited. Patient-ventilator synchrony is crucial to optimize drug delivery. The use of heliox as a driving gas may enhance the delivery of albuterol to target airways. When no measurable drop in airway resistance occurs, other causes of elevated airway resistance such as a kinked or plugged endotracheal tube should be excluded. Bronchodilator nonresponders should also be considered for a drug holiday. Data from randomized controlled trials are needed to determine the effects of bronchodilators in intubated patients, and to provide evidence for or against usual clinical recommendations for bronchodilator use.205

Other Considerations

Rarely, the above strategies are unable to stabilize the patient on the ventilator. In these situations other therapies are available. Halothane and enflurane are general anesthetic bronchodilators that acutely reduce Ppk and PaCO2.206,207 These agents cause myocardial depression, arterial vasodilation, and arrhythmias, and their benefit does not last after drug discontinuation. Heliox delivered through the ventilator circuit may decrease Ppk and PaCO2.208 However, safe use of heliox requires significant institutional expertise and careful planning. Flow meters (which are gas-density dependent) must be recalibrated for heliox, and a spirometer should be placed on the expiratory port of the ventilator during heliox administration to measure tidal volume. A trial of heliox use in a lung model is recommended prior to patient use.

Strategies to mobilize mucus such as chest physiotherapy or treatment with mucolytics or expectorants have not been proven effective in controlled trials. Bronchoalveolar lavage using either saline or acetylcysteine may be useful in nonintubated patients.209–211 This procedure is theoretically risky because the bronchoscope increases expiratory airway resistance and may worsen bronchospasm and PEEPi.

Extubation

Patients with sudden-onset asthma may respond quickly to bronchodilators. More often 24 to 48 hours of bronchodilator/anti-inflammatory therapy is required before patients are ready for extubation. As a general rule, we perform a spontaneous breathing trial once PaCO2 normalizes at a minute ventilation that does not cause significant DHI, airway resistive pressure is <20 cm H2O, mental status is intact, and significant muscle weakness is not seen. Depending on clinical status, we extubate within 30 to 120 minutes (as the endotracheal tube itself may perpetuate bronchospasm). Shortly after extubation, we administer albuterol and ipratropium bromide by nebulization. Observation in the ICU is recommended for an additional 12 to 24 hours. During this period, the focus can switch to safe transfer to a medical ward and maximizing outpatient management.