A Cardiovascular Reflexes

The cardiovascular responses to noxious airway manipulation are initiated by proprioceptors responding to tissue irritation in the supraglottic region and in the trachea.⁷ Located in close proximity to the airway mucosa, these proprioceptors consist of mechanoreceptors with small-diameter myelinated fibers, slowly-adapting stretch receptors with large-diameter myelinated fibers, and polymodal endings of nonmyelinated nerve fibers.⁸ (The superficial location of these proprioceptors and their nerves explains why topical local anesthesia of the airway is such an effective means of blunting cardiovascular responses to airway interventions.) The glossopharyngeal and vagal afferent nerves transmit these impulses to the brainstem, which, in turn, causes widespread autonomic activation through the sympathetic and parasympathetic nervous systems. Bradycardia, often elicited in infants and small children during laryngoscopy or intubation, is the autonomic equivalent of the laryngospasm response. Although seen only rarely in adults, this reflex results from an increase in vagal tone at the sinoatrial node and is virtually a monosynaptic response to a noxious stimulus in the airway.

In adults and adolescents, the more common response to airway manipulation is hypertension (HTN) and tachycardia mediated by the cardioaccelerator nerves and sympathetic chain ganglia. This response includes widespread release of norepinephrine from adrenergic nerve terminals and secretion of epinephrine from the adrenal medulla.⁹ Some of the hypertensive response to endotracheal intubation also results from activation of the renin-angiotensin system, including release of renin from the renal juxtaglomerular apparatus, which is innervated by β-adrenergic nerve terminals.

In addition to activation of the autonomic nervous system, laryngoscopy and endotracheal intubation result in stimulation of the central nervous system, as evidenced by increases in electroencephalographic (EEG) activity, cerebral metabolic rate, and cerebral blood flow (CBF).¹⁰ In patients with compromised intracranial compliance, the increase in CBF may result in elevated intracranial pressure (ICP), which, in turn, may result in herniation of brain contents and severe neurologic compromise.

The effects of endotracheal intubation on the pulmonary vasculature are probably less well understood than the responses elicited in the systemic circulation. They are often coupled with changes in airway reactivity associated with intubation. Acute bronchospasm or a main stem bronchial intubation results in a marked maldistribution of perfusion to poorly ventilated lung units, causing desaturation of pulmonary venous blood and subsequent reduction in systemic arterial oxygen (O₂) tension. In addition, institution of positive end-expiratory pressure (PEEP) after endotracheal intubation causes a reduction in cardiac output related to impaired venous return to the left side of the heart from the pulmonary circulation. The impact of these changes can be profound in patients with severely compromised myocardial function or intravascular volume depletion.

B Intubation in the Presence of Cardiovascular Disease

Myocardial ischemia results when there is an imbalance between myocardial O₂ supply and demand. In the presence of a stable O₂ content of whole blood (the product of hemoglobin concentration and saturation, with a minor contribution from dissolved O₂), the myocardial O₂ supply is almost entirely determined by coronary blood flow and distribution, because O₂ extraction at the cellular level is at or near maximum even under resting conditions.

The chief components on the demand side of the myocardial O₂ balance equation are beat frequency or heart rate (HR) and myocardial wall tension. Of the two, increases in HR are of greatest concern, because cardiac inotropism (contractility) subserves cardiac chronotropism (rate). Not only does tachycardia increase myocardial O₂ consumption per minute at constant wall tension, but elevations in rate effectively reduce the diastolic period. Because full diastolic relaxation may be impaired, a subsequent increase in resting wall tension will impair subendocardial blood flow, thereby reducing myocardial O₂ supply. Concomitantly, the rate of intraventricular pressure development in systole (dP/dT), a measure of myocardial contractility and another determinant of myocardial O₂ demand, will also increase.

It follows, then, that neuroendocrine responses to airway manipulation resulting in tachycardia and HTN may result in a variety of complications in patients with cardiac disease, myocardial ischemia chief among them. This set of circumstances is responsible for episodes of ischemic electrocardiographic ST-segment depression and increased pulmonary artery diastolic blood pressure (BP) that may be seen when intubation is performed in patients with arteriosclerosis; occasionally, these episodes presage the occurrence of a perioperative myocardial infarction.² However, short ischemic episodes (<10 minutes) evidenced by electrocardiographic ST-segment depression, such as those that may be experienced only during airway manipulation, have not been shown to correlate with postoperative myocardial infarction. In contrast, ST-segment changes of a single duration lasting longer than 20 minutes (mean SD 20 ± 30 minutes) or cumulative durations of longer than 1 hour (mean SD 1 ± 2 hours) do seem to be an important factor associated with adverse perioperative cardiac outcomes.^11,12

Patients with aneurysmal disease of the cerebral and aortic circulation may also be at particular risk of complications related to a sudden increase in BP during airway instrumentation. Laplace’s law defines the transmural wall tension of a blood vessel (the determinant of its likelihood of rupture) as the product of the pressure inside the vessel and its radius divided by the wall thickness. The presence of a thin-walled vascular aneurysm (higher transmural wall tension at baseline) combined with a sudden increase in intraluminal pressure can lead to rupture of the affected vessel and abrupt deterioration in the patient’s status. Leaking aortic aneurysms are partially tamponaded by intra-abdominal pressure but can suddenly expand into the retroperitoneal space during arterial HTN. The results are significant blood loss and additional technical problems for the surgeon trying to resect the lesion and insert a vascular prosthesis.

C Implications for Patients with Neurovascular Disease

Intracranial aneurysms and arteriovenous malformations (AVMs) often arise with a small “sentinel” hemorrhage that serves as a warning of worse things to come. During subsequent periods of elevated arterial BP, these lesions are likely to rebleed, resulting in sudden and permanent neurologic injury. Many neurosurgeons and interventional neuroradiologists attempt to stabilize cerebral aneurysms and AVMs soon after hospitalization in an effort to minimize the risk of rebleeding. This means that the patient presents for anesthesia at a time when the clot tamponading the aneurysm or AVM is particularly delicate, and a small increase in arterial transmural pressure could cause rerupture. One of the times at which this is most likely to occur is when the arterial BP and the HR are increased in response to endotracheal intubation.⁵ Therefore, neurosurgical anesthesiologists pay meticulous attention to attenuating these responses during the course of anesthetic induction and endotracheal intubation.

D Intubation in Patients with Neuropathologic Disorders

Reflex responses to endotracheal intubation are also a potential hazard to patients with compromised intracranial compliance resulting from neuropathologic processes such as intracranial mass lesions, brain edema, or acute hydrocephalus. Uncontrolled coughing can result in a marked increase in intrathoracic and intra-abdominal pressure that, in turn, increases cerebrospinal fluid pressure and may result in impairment of cerebral perfusion. In patients with impaired cerebral autoregulation (e.g., brain trauma, cerebrovascular accidents, neoplasms), the normal tendency for CBF to remain constant over the mean BP range of 50 to 150 mm Hg is impaired. When endotracheal intubation causes an increase in arterial BP, there is a marked increase in CBF and cerebral blood volume, which in turn can cause dangerous increases in ICP.⁶ This effect is magnified by the fact that noxious stimuli, such as airway manipulation, result in increased CBF, which summates with the hypertensive BP response, occasionally causing profound increases in ICP (Fig. 7-1).

Figure 7-1 Increases in systemic arterial pressure (SAP) and intracranial pressure (ICP) in response to endotracheal intubation in a patient with a small brain tumor. Notice the minimal response to rigid laryngoscopy. There is a sustained increase in systemic arterial pressure but only a transient increase in ICP, which returns to normal as cerebrovascular autoregulation becomes operative.

(From Bedford RF: Circulatory responses to tracheal intubation. Probl Anesth 2:201, 1988.)

E Neuromuscular Blocking Drugs and Cardiovascular Responses

Neuromuscular blocking drugs (NMB) are often administered to optimize conditions for intubation. Accordingly, it is appropriate to consider the cardiovascular and cerebrovascular responses to the administration of these agents. Indeed, the hypertensive-tachycardic response to endotracheal intubation was not identified until NMB agents were introduced into clinical practice, because before that time intubation was performed only with the patient under such deep levels of anesthesia that relatively little cardiovascular response was generated.¹³

The depressor effects of benzylisoquinolinium relaxants (atracurium and mivacurium) are mediated by histamine release.¹⁴ This effect could be viewed as a potential antagonist to the pressor response to laryngoscopy and endotracheal intubation. In the case of patients at risk for intracranial HTN, however, histamine-induced cerebral vasodilation may produce increases in ICP even as the BP falls.¹⁵ By contrast, pancuronium, rocuronium, and, to a lesser extent, vecuronium may initiate a hyperdynamic cardiovascular state that can potentiate the cardiovascular responses seen after endotracheal intubation in lightly anesthetized patients. The faster onset of rocuronium (doses of up to 2 mg/kg have a 90% chance of providing perfect intubating conditions) are the reason for its current widespread use as an alternative to succinylcholine for rapid-sequence intubation (RSI) and in operations expected to last longer than 1 hour.¹⁶

Succinylcholine, or diacetylcholine, is associated with bradycardia in children, particularly when doses are repeated, but is a cardiovascular stimulant in adults. This phenomenon is often associated with activation of the EEG, and patients with brain tumors may sustain marked increases in ICP after succinylcholine administration if intracranial compliance is compromised and cerebrovascular autoregulation is impaired.¹⁷ This has been shown in dogs to be a result of increased CBF related primarily to succinylcholine-induced increases in afferent muscle spindle activity at the time of fasciculation and secondarily to an elevated arterial carbon dioxide tension from fasciculation-induced carbon dioxide production.¹⁸ The evidence to substantiate the clinical relevance of these findings is lacking, however. Whereas it has been reported that succinylcholine administered to patients with brain tumors may elevate ICP by a mean of 5 to 12 torr, cerebral perfusion pressure does not change significantly, and a negative effect on neurologic outcome has not been documented.^19,20 Additionally, this phenomenon can be prevented by pretreatment with defasciculating doses of nondepolarizing NMB drugs.¹⁹ Further, when adequate ventilation is maintained, succinylcholine administered to intubated patients being treated for intracranial HTN of various causes or to those who have suffered severe head injuries caused by blunt trauma had no effect on ICP, cerebral perfusion pressure, or CBF.^21,22 As a result, succinylcholine is still considered a first-line agent for RSI in patients with acute head injury but is ideally used after pretreatment with a nondepolarizing agent and in the presence of slight hypocapnia.

F Cardiopulmonary Consequences of Positive-Pressure Ventilation

Venous return is defined by the pressure gradient between the mean systemic pressure in the peripheral venous system and the mean right atrial pressure. An increase in mean intrathoracic pressure due to positive-pressure ventilation (PPV) may be transmitted to the thin-walled, compressible superior and inferior venae cavae, effectively elevating the downstream pressure for venous return and thereby reducing venous blood return to the right atrium. Because the left side of the heart can only pump what the right side delivers, cardiac output and subsequently arterial BP may fall with PPV. Patients with decreased intravascular volume may have an exaggerated hypotensive response as a result of this phenomenon.

Those with impaired myocardial reserve may also be sensitive to the effects of PPV. However, a failing heart may also benefit from the combined effects of decreased preload and decreased afterload. In other words, PPV, particularly PEEP or continuous positive airway pressure (CPAP), diminishes the transmural wall tension of the left heart by raising juxtacardiac pressures. One common clinical scenario is a patient who responds to intubation with a brisk increase in BP and then suddenly develops acute hypotension as PPV is instituted. In such a situation, volume expansion, positional changes, and judicious use of α-adrenergic agents such as phenylephrine may be needed.

It should also be noted that both hypoxemia and hypercapnia lead to a stress-induced catecholamine response, which may mask other potential causes of hypotension. This becomes readily apparent only after intubation in critically ill patients. Prophylactic volume expansion and the immediate availability of vasoactive infusions decrease severe hemodynamic collapse in this situation.²³

A Technical Considerations: Minimizing Stimulation of Airway Proprioceptors

As a general rule, cardiovascular responses to airway maneuvers can be minimized by limiting airway proprioceptor stimulation, starting with manipulation of the larynx itself. For instance, cricoid cartilage pressure with a posterior force of 4.5 kg is widely used to prevent regurgitation of gastric contents or to facilitate laryngeal visualization. In a double-blind study, cricoid pressure resulted in a significantly greater HR and BP response to endotracheal intubation than occurred in patients whose cricoid area was gently palpated.²⁴ This is a little-recognized effect of cricoid pressure that should be considered when estimating the risk-benefit ratio of this procedure in individual patients.

Laryngoscopy itself is a moderately stimulating procedure, and use of a straight blade (Miller blade) with elevation of the vagally innervated posterior aspect of the epiglottis results in significantly higher arterial BP than does use of a curved blade (Macintosh or Corazzelli–London–McCoy [CLM]).²⁵ Newer video and optical laryngoscopes, which do not require alignment of the laryngeal axes for adequate visualization of the vocal cord inlet and subsequent intubation, have the potential to minimize the pressor response to airway manipulation by reducing the amount of force needed to displace oropharyngeal tissues and limiting cervical spine motion compared to traditional laryngoscopy with a Macintosh laryngoscope blade.²⁶ Nonetheless, reports documenting this advantage are few.

Use of the Pentax-AWS video laryngoscope (Pentax, Tokyo, Japan) has been reported to attenuate the hemodynamic response to endotracheal intubation after fentanyl/propofol induction when compared to either the GlideScope (Verathon, Bothell, WA) or the Macintosh laryngoscope (Fig. 7-2).²⁷ This finding is not universal. An earlier study comparing the Pentax-AWS to Macintosh laryngoscopy reported no significant differences in systolic BP, diastolic BP, or HR after intubation, and a separate study comparing the GlideScope and Macintosh laryngoscopy also failed to find significant differences in hemodynamic values at any point in the study.²⁸ None of these studies included patients with known cardiac disease or chronic HTN, who often have exaggerated pressor responses to stimulation; the newer airway devices may have greater value among this group compared with traditional laryngoscopy.

Figure 7-2 Percentage change from baseline in heart rate (HR, left) and mean arterial pressure (MAP, right) associated with endotracheal intubation using an Airway Scope, GlideScope, or Macintosh laryngoscope. Data values are presented as mean ± standard error. *P < 0.05 compared to Macintosh group. **P < 0.01 compared to Macintosh group. BL, Baseline; IND, 1 minute after induction; INT, at intubation; 1′ through 5′, minutes after endotracheal intubation.

(From Tsai P, Chen B: Hemodynamic responses to endotracheal intubation comparing the airway scope, GlideScope, and Macintosh laryngoscopes. Internet J Anesthesiol 24(2), 2010.)

The act of passing an endotracheal tube (ETT) is far more hemodynamically stimulating than just laryngoscopy. Surprisingly, the use of a lighted intubation stylet fails to prevent hemodynamic stimulation when the ETT is advanced past the vocal cords.²⁹ Insertion of a conventional laryngeal mask airway (LMA) after induction of general anesthesia with thiopental or propofol and fentanyl has been shown to cause less cardiovascular and endocrine response than laryngoscopy or endotracheal intubation.^30–33 The LMA has the advantage of avoiding the vagally mediated infraglottic stimulation entailed by the use of a laryngoscope, thus enabling lighter levels of general anesthesia. Furthermore, because muscle relaxation is not required for airway control, spontaneously initiated ventilation is possible, with avoidance of the adverse hemodynamic consequences of PPV. In contrast, endotracheal intubation using the intubating LMA (iLMA) resulted in a hemodynamic and endocrine response similar to that resulting from direct laryngoscopy and intubation after propofol induction.³⁴ Therefore, if endotracheal intubation is necessary, there does not appear to be a hemodynamic advantage to instrumenting the airway with the iLMA.

Whatever the technique employed to manage the airway, it must be emphasized that the hypertensive-tachycardic response to intubation is a manifestation of insufficient anesthesia. Insofar as the pressor response can also be influenced by prolonged intubation time, rapid first-attempt success is also of particular importance.⁷

B Topical and Regional Anesthesia

Topical anesthesia applied to the upper airway is effective in blunting hemodynamic responses to endotracheal intubation,^35,36 but it has almost invariably proved to be less effective than systemic administration of lidocaine. During general anesthesia, rigid laryngoscopy and instillation of lidocaine solution initiate the same adverse reflexes caused by placement of an ETT (Fig. 7-3).³⁷ Furthermore, a laryngotracheal spray of lidocaine solution may, in itself, produce profound cardiovascular stimulation in adults, and in children it may produce the same sort of bradycardic response associated with endotracheal intubation.³⁸ If topical lidocaine is administered to the upper airway, there should be an intervening period of at least 2 minutes to allow initiation of anesthetic effect before airway instrumentation begins.³⁹

Figure 7-3 Mean arterial pressure (MAP) response to endotracheal intubation after either intravenous (IV) or intratracheal (LTA) lidocaine instillation.

(From Hamill JF, Bedford RF, Weaver DC, Colohan AR: Lidocaine before endotracheal intubation: IV or laryngotracheal? Anesthesiology 55:578, 1981.)

Excellent topical anesthesia of the airway obtained before awake flexible fiberoptic intubation was responsible for reports suggesting that there was less cardiovascular stimulation after this procedure than after intubation with a rigid laryngoscope.⁴⁰ Later studies performed with patients under general anesthesia demonstrated no difference between the two modes of intubation with regard to hemodynamic impact, probably because the more profound stimulus results from placement of the ETT below the level of the glottis.^41–44

Increasing the concentration of lidocaine used, and thus the total dose, also does not appear to mitigate this effect, although it may improve intubating conditions during awake flexible fiberoptic intubation.^45,46 Although both 2% and 4% lidocaine administered through an epidural catheter in the working channel of the flexible fiberoptic bronchoscope by a “spray-as-you-go” technique provided similar intubating conditions and hemodynamic profiles, the former resulted in a smaller overall dose, lower plasma levels, and therefore less chance for toxicity reactions.⁴⁶ Lower concentrations of lidocaine (1%) provided lower plasma levels and similar hemodynamics but appeared to provide less optimal intubating conditions than atomized 2% lidocaine when used for topical anesthesia before airway manipulation.⁴⁵

In contrast to topical anesthesia of the airway, which appears to provide inconsistent benefit, regional nerve blocks involving the sensory pathways from the airway prevent hemodynamic responses to intubation. The superior laryngeal nerve (SLN) innervates the superior surface of the larynx, and the glossopharyngeal nerve innervates the oropharynx. Depositing local anesthetic on each cornu of the hyoid bone can block the SLN. Blockade of the glossopharyngeal nerve at the tonsillar pillars (sensory distribution above the level of the epiglottis) potentiates this effect by decreasing the stimulus of laryngoscopy.⁴⁷ The inferior surfaces of the larynx and trachea require topical anesthesia, however, because they are innervated by the recurrent laryngeal nerve and the vagus, which cannot be directly blocked. With the preceding combination, awake patients exhibit little response as the ETT is inserted.

Instillation of lidocaine via an ETT to prevent alterations in cerebrovascular hemodynamics in patients with severe head injury may be of some benefit. A dose of 1.7 mg/kg lidocaine instilled at body temperature given slowly (1 mL/sec) through a fine tube advanced to the end of the ETT but not in contact with the tracheal mucosa was reported to be efficacious in half of the patients treated.⁴⁸

C Inhalational Anesthetics

Defining the anesthetic dose requirement for effectively blocking (or even blunting) hemodynamic and ICP responses to endotracheal intubation has remained an elusive goal. Airway maneuvers are typically brief interventions that produce short-lived responses during a dynamic perioperative period, with drug concentrations rapidly fluctuating both in blood and at effect sites. Agents that are capable of preventing responses may also produce profound cardiovascular depression before and after the stimulation of endotracheal intubation. Accordingly, there are relatively few well-controlled dose-response studies, and those that are available often give information that is not useful for the clinical anesthesiologist.

For inhalational anesthetics, endotracheal intubation using doses in the range of the minimum alveolar concentration (1 MAC) resulted in marked cardiovascular stimulation during anesthesia with nitrous oxide (N₂O) supplemented with either halothane or morphine.⁴⁹ It should not be surprising that 1 MAC is insufficient, because it is known that approximately 1.5 to 1.6 MAC is needed to block the adrenergic and cardiovascular responses to a simple surgical skin incision (MAC-BAR).⁵⁰ The dose of anesthetic required to prevent coughing during endotracheal intubation with sevoflurane may exceed MAC by a factor of 2.86 in adults,⁵¹ although this factor appears to be close to 1.3 in children.⁵²

Accordingly, it appears that the dose of volatile anesthetic required to block the cardiovascular response to endotracheal intubation must be inordinately high, resulting in profound cardiovascular depression before endotracheal intubation.⁵³ From a cerebrovascular viewpoint, this approach is totally impractical, because high doses of volatile anesthetics cause cerebral vasodilation and marked increases in ICP in patients with compromised intracranial compliance. Furthermore, from a cardiovascular point of view, the arterial hypotension and reduced cerebral perfusion pressure before intubation would be entirely unacceptable for patients with cerebrovascular disease or brain injury.

D Intravenous Agents

Propofol, barbiturates, and benzodiazepines are all associated with profound hypotension at doses that suppress the hemodynamic and ICP responses to intubation.^54–56 In the case of etomidate, the effective dose for blocking the cardiovascular response to intubation can be identified by a burst-suppression pattern on the cortical surface EEG, indicating fairly deep cerebral depression.⁵⁷ Because etomidate supports BP at such deep levels of anesthesia, it is probably the only contemporary agent that, by itself, can achieve suppression of cardiovascular responses without first producing undue arterial hypotension and compromise of coronary and cerebral perfusion.

Because it is clinically impractical to achieve sufficient anesthetic depth for preventing a hyperdynamic response to intubation solely with an intravenous (IV) or inhalational agent (etomidate excepted), a wide variety of anesthetic drug combinations, adjuvants, or both have been used in attempts to potentiate anesthetic effects while minimizing hemodynamic depression.

Opioids are the adjuvants most commonly administered in addition to other IV or inhaled agents to facilitate induction of anesthesia and subsequent airway manipulation. Their use in this capacity relates to their historical use as part of a N₂O-narcotic anesthetic often used in patients with marginal cardiac reserve. For example, Bennet and Stanley compared the cardiovascular responses after administration of N₂O-morphine 0.4 mg/kg versus N₂O-fentanyl 4 µg/kg 10 minutes before intubation. The HR, cardiac output, and systolic and mean BP were reduced compared to baseline and remained unaffected by intubation in the N₂O-fentanyl group, but these parameters were all significantly elevated compared with preanesthetic controls in the N₂O–morphine group.⁵⁸ Whereas the assumed potency of fentanyl in this study was 100 times that of morphine, the lack of effect of morphine suggests that, with respect to suppression of pressor responses to laryngotracheal manipulation, fentanyl is more than 100 times as potent.

As reported by Bennett and Stanley⁵⁸ and later by other investigators,⁵⁹ fentanyl may not achieve its peak central nervous system effect until 10 minutes after bolus IV injection. Fentanyl appears to provide blunting of hemodynamic responses in a graded manner: 2 µg/kg IV given several minutes before induction only partially prevented HTN and tachycardia during an RSI with thiopental and succinylcholine. In this situation, 6 µg/kg was considerably more effective.⁶⁰ Chen and coworkers reported almost complete suppression of hemodynamic response to intubation with both 11 and 15 µg/kg of IV fentanyl, whereas higher IV doses (30 to 75 µg/kg) allowed only a very occasional response to intubation.⁶¹

In doses that prevent hemodynamic response to intubation, however, fentanyl is not a short-acting agent, and the risk of prolonged postoperative respiratory depression must be weighed against the advantages of perioperative cardiovascular stability. With this risk in mind, it has been observed that pretreatment with 2 µg/kg IV fentanyl given 10 minutes before intubation during an infusion of propofol sufficient to reduce the Bispectral Index Score to 45 prevented a significant increase in HR or BP compared with awake preanesthetic values.¹⁰ Similar results were observed when intubation was performed after administration of fentanyl, 2 µg/kg, and propofol bolus doses of 2.0 to 3.5 mg/kg.¹⁰

Fentanyl and propofol require 6.4 and 2.9 minutes, respectively, to achieve effect-site equilibrium after IV bolus administration.¹⁰ Therefore, the common practice of administering 1 to 2 mL (50 to 100 µg) just before or almost simultaneously with other induction medications would not be expected to have any effect based on inadequate dose and inappropriate timing of administration. Rather, this may provide a more plausible explanation for hypotension during the minutes-long quiescent period between endotracheal intubation and actual surgical incision. It is strongly recommended that laryngoscopy and intubation be timed to coincide with the peak effect of these agents.

Opioids with shorter onset and offset times have some advantages over fentanyl for modulating circulatory responses to intubation. Alfentanil has a smaller steady-state distribution volume and shorter terminal elimination half-life than fentanyl.⁶² Ausems and colleagues demonstrated that an alfentanil plasma concentration of 600 ng/mL effectively prevented hemodynamic responses to intubation during induction of N₂O anesthesia.⁶³ This was achieved by a 30-second infusion of alfentanil at 150 µg/kg. During this induction period, N₂O and succinylcholine were also administered. Only 5 of the 35 patients studied sustained an increase in HR or BP greater than 15% above preinduction values.

Remifentanil has been found to be highly effective in preventing hemodynamic responses to intubation, albeit always with the cost of impressive bradycardia or hypotension, or both, before and after airway manipulation.⁶⁴ Many studies have used vagolytic agents to avoid bradycardia, at the risk of an elevated HR response after intubation. Remifentanil’s half-time for equilibration between blood and effect site is 1.3 minutes,⁶⁵ and it has a brief half-life of 3 to 5 minutes due to hydrolysis by tissue and blood esterases.⁶⁶ Typical remifentanil infusion rates used for blunting hemodynamic responses are 0.25 to 1.0 µg/kg/min in association with cautious propofol administration and nondepolarizing neuromuscular blockade.⁶⁷ For RSI with thiopental and succinylcholine, the optimal dose of remifentanil appears to be 1.0 µg/kg administered over 30 seconds, with laryngoscopy performed 1 minute after induction. A bolus dose of 1.25 µg/kg was associated with unsatisfactory bradycardia, whereas 0.5 µg/kg resulted in excessive cardiovascular stimulation.⁶⁸ This dosing recommendation is supported by another report that found remifentanil 1 µg/kg given over 30 seconds, followed by thiopental 5 mg/kg and rocuronium 1 mg/kg 100 seconds later, was more effective than lidocaine and esmolol in attenuating the hemodynamic response to RSI.⁶⁹

IV lidocaine may also blunt hemodynamic and cerebrovascular responses to intubation. When given in a bolus of 1.5 mg/kg IV, it adds approximately 0.3 MAC of anesthetic potency.⁷⁰ Significant reductions in hemodynamic response to endotracheal intubation have been noted when lidocaine (3 mg/kg) was used as an adjunct to high-dose fentanyl anesthesia,⁷¹ as well as during other light anesthetic techniques, such as thiopental-N₂O-O₂.⁷² However, smaller doses of lidocaine (1.5 mg/kg) have not been consistently reported to be effective in reducing the hemodynamic response to laryngoscopy and endotracheal intubation.^73,74

The general anesthetic properties of lidocaine tend to reduce cerebral metabolic rate for O₂ and CBF, thus lowering ICP in patients with compromised intracranial compliance.⁷⁵ Theoretically, these properties of lidocaine might be exploited to mitigate rises in ICP during airway manipulation in those patients with acute intracranial pathology or compromised intracranial compliance. However, only a single human study has been reported specifically evaluating the ability of IV lidocaine to blunt intubation-related elevations in ICP. Bedford and colleagues compared 1.5 mg/kg IV lidocaine with placebo in 20 patients diagnosed with brain tumor. When administered 2 minutes before intubation, lidocaine failed to prevent a rise in ICP from the preanesthesia baseline, although the increase was more modest than that observed with the placebo (−12.1 mm Hg; 95% confidence interval, −22.8 to −1.4 mm Hg; P = 0.03).⁷⁶ This dearth of direct benefit was underscored by a systematic review that also failed to identify any evidence that pretreatment with IV lidocaine before RSI consistently reduced ICP or positively affected neurologic outcome.⁷⁷ This review is now more than a decade old, but because no new direct evidence has been published in the interim, its conclusion remains valid.

With regard to the patient at risk for intracranial HTN, it is important that agents used to control cardiovascular responses to intubation also have a minimal adverse impact on ICP. Agents that act as cerebral vasodilators, such as volatile anesthetics, nitroglycerin, nitroprusside, or hydralazine, are generally avoided if there is a serious risk of intracranial HTN.

E Nonanesthetic Adjuvant Agents

A final means for modifying the cardiovascular responses to endotracheal intubation is prophylactic administration of vasoactive substances that directly affect the cardiovascular system. This approach was introduced in 1960 by DeVault and associates, who found that pretreatment with phentolamine, 5 mg IV, prevented the hypertensive-tachycardic response to endotracheal intubation during a light barbiturate-succinylcholine anesthetic technique.⁷⁸ Since then, a large number of articles have appeared advocating the use of various vasodilators and adrenergic blocking agents as pretreatment before endotracheal intubation, including diltiazem, verapamil, and nicardipine^79–82; hydralazine⁸³; nitroprusside⁸⁴; nitroglycerin⁸⁵; labetalol⁸⁶; esmolol^80,87–89; and clonidine.^90,91 Virtually all of these agents appear to be somewhat effective when compared to placebo, particularly when used in high doses.

Esmolol is the best studied of the group. In a large, multicenter, placebo-controlled trial, esmolol at doses of 100 or 200 mg suppressed the hemodynamic response to endotracheal intubation, particularly when combined with a moderate-dose opiate.⁸⁷ However, esmolol doses of 200 mg were associated with a doubling of the incidence of hypotension compared to placebo. In another study, smaller doses of esmolol (1 mg/kg) had no effect on the hemodynamic response to laryngoscopy and intubation compared to placebo.⁸⁰ Most recently, the combination of lidocaine (1.5 mg/kg) and esmolol at a dose of 1 mg/kg effectively attenuated the pressor response to intubation but was not as effective as 1 µg/kg remifentanil.⁶⁹ Currently, the optimal use of any of these agents is undefined, although their use as adjuncts to RSI is reasonable taking into account evidence-based dosing recommendations for the situation.

A Upper Airway Reflexes

Because the upper airway protects the respiratory gas exchange surface from noxious substances, it is appropriate that the nose, mouth, pharynx, larynx, trachea, and carina have an abundance of sensory nerve endings and brisk motor responses. Anesthesiologists are especially familiar with the glottic closure reflex (laryngospasm), which is invariably encountered early in their training. The sneeze, cough, and swallow reflexes are equally important upper airway reflexes.

Afferent pathways for laryngospasm and the cardiovascular responses to endotracheal intubation are initiated by the glossopharyngeal nerve when stimulation occurs superior to the anterior surface of the epiglottis and by the vagus nerve when stimulation occurs from the level of the posterior epiglottis down into the lower airway. Because the laryngeal closure reflex is mediated by vagal efferents to the glottis, it is virtually a monosynaptic response, occurring primarily when a patient is lightly anesthetized as vagally innervated sensory endings in the upper airway are stimulated and conscious respiratory efforts cannot override the reflex.

B Dead Space

Patients with severe chronic lung disease may find it easier to breathe after intubation or a tracheostomy. The improvement is most likely due to reduced dead space. The normal extrathoracic anatomic dead space, based on cadaver measurements, is between 70 and 75 mL.⁹² The exact volume (V) of the ETT is easily calculated as that of a cylinder using the formula V = r²l, where r is the radius of the tube and l is the length. For example, an ETT that has an 8-mm inner diameter (ID) of 25 cm and a length of 25 cm has a volume of 12.6 mL. Intubation would therefore result in a reduction in dead space of approximately 60 mL.

Tracheostomy tubes are shorter than oral ETTs and have an even smaller dead space, although the difference as a proportion of tidal volume (V_t) is negligible. In normal individuals, the reduction in dead space with insertion of a tracheostomy tube is negligible compared with the normal V_t, so there is little benefit. In a patient with severe restrictive lung disease such as in end-stage kyphoscoliosis, V_t may be as low as 100 mL and intubation can confer a major benefit. Similarly, patients with emphysema who are changed from mouth breathing to tracheostomy demonstrate a reduction in minute volume required and a decrease in total body O₂ consumption, presumably due to a decreased work of breathing.⁹³ Most likely, the decreased volume required more than compensates for the slight increase in resistance.

The decreases in dead space described here refer only to the volume of the ETT alone and are applicable only during T-piece breathing. Any extensions or Y-pieces added to the breathing circuit and attached to the ETT must also be added to the total.

C Upper Airway Resistance

Anesthesiologists are well aware that in most anesthetized patients, adequate ventilation can be maintained with an ETT as small as 6 mm ID in place. Intensivists caring for a patient with respiratory failure often insist that an ETT must have a minimum ID of 8 mm. These tube sizes are each appropriate for the clinical situations described. The high resistance of the 6 mm ETT is inconsequential for the low minute ventilation required under general anesthesia, but the high flow rates required for a patient with respiratory failure may render the resistance of a small ETT prohibitive.

The ETT creates a mechanical burden for a spontaneously breathing patient in the form of a fixed upper airway resistance; that is, it decreases airway caliber and increases resistance to breathing. Gas flow across an ETT is determined by the pressure difference across the tube and the resistance of the ETT. Gas flows whenever there is a pressure difference across the ETT, whether it is caused by subatmospheric pressure generated during spontaneous breathing or by positive pressure generated from a mechanical ventilator.

The apparent resistance of an ETT is influenced by the shape of the tube and by two types of friction: the friction among the gas molecules and the friction between the tube wall and the gas molecules.^94,95 Irregular surfaces created by secretions or by ridges from wire reinforcement may create greater friction and greater resistance.⁹⁶ ETTs and tracheostomies have a higher resistance than the normal upper respiratory tract.^97,98

The relationship between pressure difference and flow rate depends on the nature of the flow: laminar, turbulent, or a mixture of the two. In an ETT, turbulent flow predominates. During turbulent flow, the measured resistance is not a constant but varies with the flow rate, becoming markedly higher at high flow rates. Instead of the laminar flow relationship of pressure being directly proportional to flow, the pressure required to move the gas through an ETT with turbulent flow is proportional to the square of the flow. The relationship can therefore be described by a parabolic curve, as in Figure 7-4. The apparent resistance of a tube is proportional to the fourth power of the radius during laminar flow (Poiseuille’s law) but to the fifth power during turbulent flow. Assuming turbulent flow, the relative resistance of a 6-mm ETT versus an 8-mm ETT is 4⁵/3⁵, or 4.2 times as great. However, because flow patterns are not entirely predictable, the exact respiratory pressure-flow relationships may not be predictable without empiric determination. Such determinations are depicted in Figure 7-4. The slope of the pressure-flow graph is the apparent resistance. The parabolic shape of the graphs demonstrates the primarily turbulent nature of the flow through an ETT.

Figure 7-4 Pressure drop across endotracheal tubes of various sizes at flow rates ranging from 0 to 100 L/min. Note the wide disparity between 6- and 7-mm tubes as flow rate increases to the range typically seen in patients with respiratory failure.

(Adapted from Nunn JF: Applied respiratory physiology, London, 1987, Butterworths.)

Although the resistance of the ETT may be several-fold greater than the resistance of the normal human upper airway, this is of relatively little consequence at low minute ventilation.⁹⁹ With a typical peak inspiratory flow of 25 to 30 L/min, approximately 0.5 cm H₂O pressure must be generated to overcome the resistance of the upper respiratory tract. This represents about 10% of the total work of breathing. Even a doubling or tripling of that resistance by placement of an ETT does not result in a clinically worrisome increase in the total work of breathing.¹⁰⁰

As flow rates increase, however, flow becomes more turbulent and tube resistance may become a problem. For flow rates greater than 15 L/min, flow through any tube smaller than 10 mm ID becomes turbulent.¹⁰¹ At the high flows required by patients in respiratory failure, the resistance of smaller tubes becomes prohibitive.¹⁰² At the time of weaning patients with respiratory failure from mechanical ventilation, the importance of tube size often becomes a critical factor, with a common question being whether to change a 7-mm to a larger size. Note in Figure 7-4 that the pressure drop between 7-mm and 9-mm tubes is minimal at low inspiratory flows, but is considerable at flows of 60 L/min and higher.

Theoretically, the patient’s native airway should have less resistance than any size ETT. However, a patient who has been intubated for an extended period may not have a normal upper airway. Indeed, some evidence suggests that work of breathing may actually increase after extubation, perhaps due to high upper airway resistance.^103,104 Therefore, if the patient is close to successful weaning, a reasonable approach may be to attempt extubation rather than to change ETTs, recognizing that the need for reintubation is a possibility. Alternatively, pressure support ventilation can be used to compensate for the added work of breathing through the smaller tube until extubation is warranted.¹⁰⁵

Tracheostomy tubes have lower resistance than ETTs of comparable diameters because they are shorter. However, there is little, if any, difference in the work of breathing imposed by fresh tracheostomy tubes and ETTs of comparable ID.¹⁰⁶ On the other hand, tracheostomy does appear to decrease the work of breathing in patients who have undergone prolonged intubation and mechanical ventilation. This paradox may be explained by a reduction in the ID of an ETT over time, perhaps as a result of inspissated secretions or conformational changes.¹⁰⁷ This size reduction may explain the observation that patients being weaned from mechanical ventilation are sometimes more rapidly weaned after a tracheostomy is performed,¹⁰⁸ although it may also reflect the increased comfort of clinicians in discontinuing ventilatory support after the airway is secured.

D Lower Airway Resistance

Bronchospasm after induction of anesthesia is a relatively uncommon but well-recognized event and is likely related to a reflex response to endotracheal intubation. Several studies provide some evidence regarding the frequency of bronchospasm. Tiret and associates studied complications at the time of induction of anesthesia and noted that bronchospasm accounted for 5.3% of fatal or near-fatal peri-induction complications.¹⁰⁹ The largest population was reported by Olsson, who found 246 cases of bronchospasm to have occurred out of a total of 136,929 anesthesia inductions, for an incidence of 1.7 per 1000.¹¹⁰ However, the exact incidence undoubtedly depends substantially on the patient population.

The incidence of postintubation bronchospasm may be decreasing because of the increasing use of propofol as an induction agent (propofol being more effective than thiopental at preventing this complication). However, ventilation problems combined with hypoxia due to acute bronchospasm still represent important sentinel anesthesia events.¹¹¹

Whereas the incidence of overt clinical bronchospasm is low, a reflex increase in airway resistance may occur much more often. Receptors in the larynx and upper trachea may cause large airway constriction distal to the tube, which in turn may extend to the smaller peripheral airways.¹¹² Support for this hypothesis comes from the work of Gal, who found an increase in lower airway resistance in a series of volunteers whose tracheas were intubated after topical anesthesia (Fig. 7-5).¹¹³

Figure 7-5 Pressure and flow curves (labeled A and B, respectively) generated during a burst of three successive coughs (C) by a volunteer before (control) and after endotracheal intubation. Notice that the flows and pressures generated are only modestly diminished after intubation.

(From Gal TJ: How does tracheal intubation alter respiratory mechanics. Probl Anesth 2:191, 1988.)

Bronchoconstriction also occurs after endotracheal intubation of normal subjects who have received thiopental/narcotic anesthesia.¹¹⁴ In a series of patients pretreated before anesthesia with either a β-adrenergic agonist (albuterol) or an inhaled anticholinergic agent (ipratropium bromide), measured airway resistance after intubation was markedly lower compared with placebo treatment (Fig. 7-6).

Figure 7-6 Lung resistance (R_L) at 2, 5, and 15 minutes after intubation in patients pretreated with either a placebo, the β-adrenergic agonist albuterol, or the anticholinergic drug ipratropium bromide. Each of the drugs markedly diminished lung resistance for longer than 15 minutes after endotracheal intubation under thiopental-narcotic anesthesia.

(From Kil HK, Rooke GA, Ryan-Dykes MA, et al: Effect of prophylactic bronchodilator treatment on lung resistance after tracheal intubation. Anesthesiology 81:43, 1994.)

Increases in airway resistance may result from changes in intrinsic smooth muscle tone, airway edema, or intraluminal secretions. These factors are, in turn, controlled by a series of intracellular and extracellular events, including neural and hormonal factors. Rapid changes in airway caliber after airway instrumentation are thought to result largely from parasympathetic nervous system activation of airway smooth muscle.^115,116 Cholinergic innervation predominates in the larger central airways, with efferent nerves arising in the vagal nuclei of the brain stem and synapsing with ganglia in the airway walls. Postganglionic parasympathetic nerves release acetylcholine, activating muscarinic receptors on airway smooth muscle that lead to smooth muscle constriction. Such responses can be blocked via muscarinic blockade, using either systemic or inhaled anticholinergic agents.

Endotracheal intubation also may induce bronchospasm by causing coughing. A cough reduces lung volume, which in turn markedly increases bronchoconstriction in response to a stimulus.¹¹⁷ In the patient with known reactive airways, prevention of coughing at the time of endotracheal intubation by use of either a deep level of anesthetic or a muscle relaxant may help to minimize the likelihood of bronchospasm.

E Endotracheal Tube Resistance and Exhalation

In normal patients breathing at moderately elevated minute ventilation, exhalation is usually completed well before the next inhalation begins. By contrast, patients with obstructive airways disease may not complete full exhalation before the start of the next inhalation. In other words, inhalation begins before exhalation to functional residual capacity (FRC), resulting in persistent positive pressure in the alveoli. This phenomenon has been called auto-PEEP or dynamic hyperinflation, and it results in air trapping, elevated intrathoracic pressure, and hemodynamic compromise.¹¹⁸

Auto-PEEP most commonly occurs in patients with obstructive lung disease and high minute ventilation, but it may also rarely occur in patients with relatively normal airways who are ventilated at very high minute ventilation. This has been observed in patients with burns or sepsis, who may require as much as 30 to 40 L of minute ventilation. Under these circumstances, the resistance of the ETT may limit expiratory flow so that full exhalation does not occur. This has been demonstrated experimentally, with the magnitude of the auto-PEEP correlating directly with the resistance of the ETT.¹¹⁹ Among patients under anesthesia, major resistance to exhalation caused by the ETT is of no consequence in routine cases and is only rarely seen in critically ill patients. However, low levels of auto-PEEP due to tube resistance probably occur frequently in patients with high minute ventilation¹⁰⁸ and during single-lung ventilation via a double-lumen ETT.

F Functional Residual Capacity

The effect of endotracheal intubation on FRC has been a subject of considerable controversy. Intensivists are well aware of patients recovering from respiratory failure in whom oxygenation improved after extubation. The improvement has been attributed to “physiologic PEEP”—the presumption that a small positive pressure is normally created by the glottis and that this leads to breathing at a higher lung volume. The assumption is that an ETT removes the glottic barrier and may, therefore, lower lung volume. However, the existence of positive intratracheal pressure has never been documented, and in a study of volunteers who underwent awake intubation, no consistent change in FRC could be measured.^120–122

By contrast, a different conclusion was reached in a series of patients who were studied just before and after extubation following recovery from respiratory failure. In this situation, both FRC and arterial oxygen tension (PaO₂) were found to increase after extubation, supporting the concept that the presence of an ETT decreases FRC.¹²³ Resolution of these disparate results was suggested by a rabbit study in which normal rabbits did not demonstrate a difference in oxygenation or tracheal pressure after intubation, but after respiratory failure was induced, endotracheal intubation worsened oxygenation.¹²⁴ These results suggest that the rabbits compensated for respiratory failure by using glottic closure to maintain a positive intratracheal pressure and that the effect of an ETT on FRC depends on that underlying respiratory state.

G Cough

Although it is widely recognized that cough efficiency is reduced whenever an ETT is in place, it is a common observation that a disconnected ETT is likely to produce a plug of sputum whenever the patient is stimulated to cough. In awake intubation volunteers, peak airway flow was reduced but was still adequate to enable secretion clearance.¹¹³ However, the ETT prevented collapse of the trachea by acting as a stent. Therefore, although secretions could be moved to the central airways, the ETT prevented maximum efficiency of expectoration. Large airway collapse is important for producing maximum force against secretions, and this explains why moving secretions from the trachea out through the ETT often requires the use of a suction catheter.

H Humidification of Gases

Under normal circumstances, the upper airway warms, humidifies, and filters 7000 to 10,000 L of inspired air daily, adding up to 1 L of moisture to the gases. When the upper airway is bypassed by intubation, the gas must be warmed and humidified in the trachea if it is not adequately humidified before inhalation. In an anesthetized patient breathing dry gases, up to 10% of the average metabolic rate may be required to perform these tasks.¹²⁵ Delivery of cool, dry gases may also have a significant effect on mucociliary transport, a critical defense mechanism of the respiratory tract. Inhalation of unconditioned gas rapidly leads to abnormal mucosal ciliary motion, with subsequent encrustation and inspissation of tracheal secretions.^126,127 These changes occur as early as 30 minutes after intubation and, theoretically, may lead to an increase in postoperative complications in patients with limited chest excursion. Accordingly, assurance of adequate gas conditioning should be standard in all but very brief endotracheal intubations.

A Preventing Upper Airway Responses

Cough and laryngospasm in response to intubation appear to be sound protective reflexes. Under most circumstances, the body needs to prevent further intrusion by a foreign body and to expel it from the airway. However, these responses can be troublesome during induction of anesthesia or at the time of extubation. Cough can lead to bronchospasm as lung volume is reduced, and it can also result in desaturation as the lung volume drops to residual volume. Laryngospasm may result in life-threatening abnormalities of blood gases. Consequently, anesthesiologists routinely try to prevent these responses with the use of medications delivered topically, via inhalation, or intravenously.

Inhibition of upper airway reflexes can certainly be accomplished by performing endotracheal intubation after the administration of NMB agents. However, both laryngeal and tracheal reflexes are difficult to inhibit by deep levels of general anesthesia alone.¹²⁸ When circumstances preclude the use of NMB agents, the clinician must give consideration to how best to prevent discomfort, gagging, coughing, and laryngospasm during endotracheal intubation: avoidance of endotracheal intubation, use of regional and topical anesthesia, very deep general anesthesia, or a combination of all modalities.

2 Regional and Topical Anesthesia

The surfaces of the mouth and nose are easily anesthetized with topical anesthetic sprays or gels. Lidocaine is equally effective as cocaine and less toxic; it can be combined with a vasoconstrictor to give equivalent intubating conditions.^130–132 Administration of an antisialagogue 30 to 60 minutes before application of the topical anesthetic results in better anesthesia as well as better intubating conditions. The lack of secretions probably minimizes dilution of the applied anesthetic and also results in better intubating conditions.

The mouth and pharynx derive their sensory innervation from the trigeminal and glossopharyngeal nerves. The supraglottic larynx derives its sensory innervation from the SLN, a branch of the vagus, and intubation can be facilitated by blocking it bilaterally.¹³³ The nerve block relies on the consistent relationship of the SLN to the lateral horns of the hyoid bone. When combined with topical anesthesia of the nose or mouth and adequate anesthesia of the infraglottic larynx, the nerve block provides excellent intubating conditions, and most patients are able to accept an ETT without cough, gag, or laryngospasm. Equal success in blunting upper airway reflexes can be achieved by careful spraying of the larynx with topical anesthesia. A nasal trumpet helps ensure that solution reaches the larynx. Topical anesthesia spares the patient the need for two injections.

The infraglottic larynx derives sensory innervation from the recurrent laryngeal nerves, which run along the posterolateral surfaces of the trachea. Again, topical anesthesia rather than nerve block is the method of choice for obtunding reflexes. Injection of several milliliters of 4% lidocaine via the cricothyroid membrane routinely results in excellent blockade of sensation.

The efficacy of topical and nerve block anesthesia at suppressing airway reflexes during intubation is evident. Several studies have documented that topical anesthesia applied preoperatively (for brief cases) or intraoperatively can suppress cough and laryngospasm at the time of extubation.¹³⁴ A randomized study of patients undergoing tonsillectomy found that the incidence of stridor or laryngospasm at the time of extubation could be reduced from 12% to 3% by application of topical lidocaine at the time of intubation.¹³⁵ The LITA endotracheal tube (Laryngotracheal Instillation of Topical Anesthetic, Sheridan Corporation, Argyle, NY) contains a small channel that can be used to spray the upper airway while an ETT is in place. When this method was used to spray the ETT before extubation, coughs were reduced by more than 60%, and the severity of the coughing was decreased.¹³⁶

The use of liquid, in the form of a lidocaine-bicarbonate mixture, rather than air to inflate the cuff of the ETT after intubation has also been reported to be effective in diminishing emergence phenomena.^137,138 Inflating the ETT cuff with 40 mg of lidocaine (2 mL of 2% solution) and then adding 3 to 7 mL of 8.4% sodium bicarbonate (NaHCO₃) until no cuff leak was present resulted in significant reductions in coughing, restlessness, and BP during emergence. In addition, sore throat complaints assessed at 15 minutes and at 1, 2, 3, and 24 hours postoperatively; postoperative dysphonia; and hoarseness after extubation were all reduced when compared to cuff inflation with air. When this technique was used, more than 50% of the original 40 mg of lidocaine still remained in solution at 2 hours after inflation of the cuff, and only about 75% was released even after 6 hours of surgery (Fig. 7-7).¹³⁷ Because standard 8.4% NaHCO₃ is a basic solution with a calculated pH of 7.8 (range 7 to 8.5), the addition of more than 2 mL of bicarbonate to the 2 mL of 2% lidocaine (calculated pH 6, range 5 to 7) already injected into the ETT cuff results in a solution with a pH 7.95 to 8.09; this leads to concern about tracheal mucosal damage from flash burn injury in the event of a cuff rupture. However, a direct comparison between solutions of 2 mL of 2% lidocaine with 8.4% versus 1.4% bicarbonate reported similar efficacy in reducing postoperative sore throat complaints and the occurrence of various emergence phenomena.¹³⁸ Therefore, in actual clinical practice, a favorable risk-benefit balance can be achieved by using the following combination in a 10-mL syringe: 5 mL 1% lidocaine, 1 mL 8.4% NaHCO₃ solution, and 4 to 5 mL of sterile diluent (J.P. Estebe, personal communication, 2010).

Figure 7-7 Percentage of lidocaine released in vitro as a function of time from an endotracheal tube (ETT) cuff filled with 40 mg lidocaine hydrochloride and additional 8.4% sodium bicarbonate solution equal to 0, 3, 4, 5, 6, or 7 mL.

(From Estebe JP, Dollo G, Le Corre P, et al: Alkalinization of intracuff lidocaine improves ETT-induced emergence phenomena. Anesth Analg 94:227–230, 2002.)

B Preventing Bronchoconstriction

Bronchoconstriction results routinely after endotracheal intubation. In healthy subjects it appears to be of moderate degree, but the exaggerated response seen in patients with hyperactive airways can be life-threatening. Prevention or treatment of this response can be achieved with the use of topical or IV agents. Inhaled anesthetic agents also inhibit the response through direct absorption by smooth muscle or inhibition of reflexes.

Bronchospasm after intubation may be cholinergically mediated. Afferent parasympathetic fibers travel to bronchial smooth muscle and then produce bronchoconstriction by stimulating the M₃ cholinergic receptors on bronchial smooth muscle. In addition, stimulation of M₂ cholinergic receptors on airway smooth muscle potentiates bronchospasm by inhibiting β-adrenergic–mediated smooth muscle relaxation.¹⁴⁵

3 Intravenous Agents

A variety of drugs have been studied for their bronchodilating properties. Although IV β-agonists clearly produce bronchodilation, there is no benefit to parenteral administration of these drugs rather than inhalational administration. Among anesthetic induction agents, considerable experimental evidence suggests that ketamine has both direct and indirect relaxant effects on airway smooth muscle through non–β-receptor mechanisms.^152–157 However, the clinical data supporting the use of ketamine for prevention or treatment of bronchospasm is largely anecdotal,¹⁵⁸ or in more rigorous trials unimpressive.^159–161 This may relate to reluctance to routinely use ketamine at high doses because of its side effects, including dysphoria and sympathetic stimulation, rather than a lack of benefit of the drug.

Propofol, midazolam, and etomidate all relax airway smooth muscle in vitro, although generally at higher site concentrations than would be used clinically.^162–166 In contrast, barbiturates may have direct bronchoconstricting effects.¹⁶⁷ Propofol may also have indirect effects on airway constriction, perhaps though inhibition of vagal tone.^157,168 Clinically, propofol has been shown to be superior to the barbiturates and to etomidate in reducing wheezing and airway resistance in both asthmatic and nonasthmatic subjects.^169–171 When asthmatics were induced with either thiopental, methohexital, or propofol at equipotent doses, none of the patients given propofol wheezed after endotracheal intubation, whereas both of the barbiturates resulted in a significant incidence of wheezing (Fig. 7-8).¹⁶⁹

Figure 7-8 Incidence of wheezing after endotracheal intubation in asthmatics when induction was performed with either an oxybarbiturate, a thiobarbiturate, or propofol (P < 0.05 for either thiobarbiturate or oxybarbiturate versus propofol).

(From Pizow R, Brown RH, Weiss YS, et al: Wheezing during induction of general anesthesia in patients with and without asthma: A randomized, blinded trial. Anesthesiology 81:1111, 1995.)

In animals, IV lidocaine has been reported to attenuate bronchoconstriction induced by a variety of experimental means.^172,173 In humans with bronchial hyperreactivity, IV lidocaine reduced the bronchoconstrictor response to histamine challenge and had an additive effect with albuterol in reducing this response.¹⁷⁴ However, a double-blind, placebo-controlled trial of IV lidocaine (1.5 mg/kg) or inhaled albuterol in asthmatics found that albuterol but not lidocaine prevented postintubation bronchoconstriction.¹⁷⁵ Although lidocaine may inhibit reflex-induced bronchospasm, it may also cause contraction of bronchial smooth muscle in the absence of reflex mechanisms. In a study of 15 asthmatics, IV lidocaine reduced airway diameter at total lung capacity assessed by computed tomography and resulted in significant decreases in FEV₁.¹⁷⁶ These untoward effects were underscored by a case report in which IV lidocaine 1.5 mg/kg administered to facilitate intubation was temporally associated with transient bronchospasm in a 17-month-old child with mild intermittent asthma.¹⁷⁷ A published best-evidence review also failed to find good evidence for use of IV lidocaine during intubation in patients with status asthmaticus.¹⁷⁸ Therefore, the evidence in support of IV lidocaine to prevent postintubation bronchospasm when used without the concomitant administration of an inhaled β-agonist is scant, and there is a potential risk of worsening airway resistance with its use. Therefore, the use of IV lidocaine for this indication cannot be endorsed.

4 Inhaled Agents

All of the volatile anesthetics have direct and perhaps indirect relaxant effects on airway smooth muscle in experimental models.^94,179–182 Although these agents have differences in potency in vitro, the clinical importance of these differences is unclear.^180,182,183 In adult patients, sevoflurane is more effective than isoflurane, desflurane, or halothane in reducing airway resistance after endotracheal intubation,^184–186 but does not prevent an increase in airway resistance after intubation of asthmatic children.¹⁸⁷ However, given the available data, sevoflurane is probably the volatile agent of choice, and desflurane is a poor choice for use in high-risk patients (Fig. 7-9).

Figure 7-9 Respiratory system resistance (percent of baseline) during maintenance anesthesia. A, Isoflurane, halothane, and sevoflurane are compared with thiopental 0.25 mg/kg/min plus 50% nitrous oxide. *P < 0.05 versus isoflurane, halothane, and thiopental. ^†P < 0.05 versus thiopental. B, Desflurane and sevoflurane are compared with thiopental 0.25 mg/kg/min. *P < 0.05 versus desflurane and thiopental.

(A, Adapted from Rooke GA, Choi JH, Bishop MJ: The effect of isoflurane, halothane, sevoflurane, and thiopental/nitrous oxide on respiratory system resistance after tracheal intubation. Anesthesiology 86:1294, 1997; B, From Goff MJ, Arain SR, Ficke DJ, et al: Absence of bronchodilation during desflurane anesthesia: A comparison to sevoflurane and thiopental. Anesthesiology 93:404, 2000.)

There are no prospective, controlled studies comparing deep inhalation anesthesia to IV induction with bronchoprotective agents such as ketamine or propofol in high-risk patients. Achieving a deep plane of anesthesia with a bronchoprotective agent before intubation is likely the most important point in preventing severe bronchospasm in high-risk patients, rather than the choice of IV versus inhalation induction techniques.

Pretreatment of patients with inhaled β₂-adrenergic agonists or an inhaled anticholinergic markedly reduced lung resistance following endotracheal intubation and should be used routinely in patients known to have bronchospasm.^114,188