Manual of Emergency Airway Management, 3rd Edition

28.Elevated Intracranial Pressure

Andy S. Jagoda

John J. Bruns Jr.

The Clinical Challenge

Elevated intracranial pressure (ICP) poses a direct threat to the viability and function of the brain. In head trauma, elevated ICP has been clearly associated with worse outcomes. The problems associated with elevated ICP may be compounded by many of the techniques and drugs used in airway management because they may cause further elevations of ICP. In addition, victims of multiple trauma may present with hypotension, thus limiting the choice of agents and techniques available. This chapter provides the basis for an understanding of the problems of increased ICP and the optimal methods of airway management in this patient group.

When increased ICP occurs as a result of an injury or medical catastrophe, the brain's ability to regulate blood flow (autoregulation) over a range of blood pressures is often lost. In general, the ICP is maintained through a mean arterial pressure (MAP) range of 80 to 180 mm Hg. When the ICP becomes elevated, autoregulation often, but not always, has been lost. In this setting, excessively high or excessively low blood pressure could aggravate brain injury by promoting cerebral edema or ischemia. Hypotension, even for a very brief period, is especially harmful, and along with hypoxia, has been shown to be an independent predictor of mortality and morbidity in patients with traumatic brain injury (TBI).

Cerebral perfusion pressure (CPP) is the driving force for blood flow to the brain. It is measured by the difference between the MAP and the ICP, expressed as the formula:

CPP = MAP - ICP

It is clear from this formula that excessive decreases in MAP, as might occur during rapid sequence intubation (RSI), would decrease CPP and contribute to cerebral ischemia. Conversely, increases in MAP, if not accompanied by equivalent increases in ICP, may be beneficial because of the increase in the driving pressure for oxygenation of brain tissue. It is generally recommended that the ICP be maintained below 20 mm Hg, the MAP between 100 to 110 mm Hg, and the CPP near 70 mm Hg. There are a number of confounding elements that may increase ICP during airway management.

Reflex Sympathetic Response to Laryngoscopy

The reflex sympathetic response to laryngoscopy (RSRL) is stimulated by the rich sensory innervation of the supraglottic larynx. Use of the laryngoscope, and particularly the attempted placement of an endotracheal tube, results in a significant afferent discharge that increases sympathetic activity to the cardiovascular system mediated through direct neuronal activity and release of catecholamines. More prolonged or aggressive attempts at laryngoscopy and intubation result in greater sympathetic nervous system stimulation. This catecholamine surge leads to increased heart rate and blood pressure, which significantly enhances cerebral blood flow (CBF) at the apparent expense of the systemic circulation through redistribution. These hemodynamic changes may contribute to increased ICP, particularly if autoregulation is impaired; therefore, it is desirable to mitigate this RSRL. Gentle intubation techniques that minimize airway stimulation and pharmacological adjuncts (e.g., beta blockade, lidocaine, synthetic opioids) have been studied to accomplish this mitigation.

Evidence is mixed regarding the use of lidocaine to blunt the hemodynamic response to laryngoscopy. Studies in patients without cardiovascular disease have failed to show effect, and other studies have shown variable results with respect to hemodynamic protection, with some appearing to demonstrate benefit and others showing none. As a result, lidocaine cannot be recommended at the present time for mitigation of the RSRL associated with emergency intubation.

The short-acting beta-blocker esmolol, in contrast, has consistently demonstrated the ability to control both heart rate and blood pressure responses to intubation. A dose of 2 mg/kg given 3 minutes before intubation has been shown to be effective. Unfortunately, in the emergency situation, the administration of a beta-blocking agent, even one that is short acting, may be problematic by causing or exacerbating hypotension in a trauma patient, or by confounding interpretation of a decrease in the blood pressure changes immediately following intubation. For these reasons, although esmolol is consistent and reliable for mitigation of RSRL in elective anesthesia, it is generally not used for this purpose for emergency intubation.

Fentanyl at doses of 3 to 5 µg/kg has also been shown to attenuate the RSRL associated with intubation. Although a full sympathetic blocking dose of fentanyl is 9 to 13 µg/kg, the recommended pretreatment dose of fentanyl for emergency RSI is 3 µg/kg and should be administered as a single pretreatment dose over 60 seconds. This technique permits effective mitigation of the RSRL, with greatly reduced chances of apnea or hypoventilation before sedation and paralysis.

Several studies have investigated the potential advantages of fiberoptic or lighted stylet intubation over direct laryngoscopy, working on the premise that these techniques minimize tracheal stimulation and thus the RSRL. Results of these studies are mixed and do not permit any conclusions recommending one technique over the other. In a controlled operating room setting, the insertion of the endotracheal tube into the trachea is more stimulating than a routine laryngoscopy.

At present, based on the best available evidence, it seems advisable to administer 3 µg/kg of fentanyl intravenously (IV) as a pretreatment agent 3 minutes before administration of the induction and neuromuscular blocking agents to mitigate the RSRL. Fentanyl should not be administered to patients with incipient or actual hypotension or to those who are dependent on sympathetic drive to maintain an adequate blood pressure for cerebral perfusion. In such cases, the ensuing hypotension may cause further central nervous system injury. In addition to pharmacological maneuvers to reduce RSRL, intubation should be performed in the gentlest manner possible, limiting both the time and intensity of laryngoscopy.

Reflex ICP Response to Laryngoscopy

Laryngoscopy may also increase the ICP by a direct reflex mechanism not mediated by sympathetic stimulation of the blood pressure or heart rate. The details of this reflex are poorly elucidated. Insertion of the laryngoscope or endotracheal tube may, therefore, further elevate ICP, even if the RSRL is blunted. It would seem desirable to blunt this ICP response to laryngoscopy in patients at risk for having elevated ICP. The literature related to the use of lidocaine to blunt ICP response to laryngoscopy and intubation is discussed in Chapter 17. In patients with elevated ICP, lidocaine should be administered as a pretreatment drug in the dose of 1.5 mg/kg IV 3 minutes before the induction agent and succinylcholine (SCh) to mitigate the ICP response to laryngoscopy and intubation.

ICP Response to Succinylcholine

SCh itself may be capable of causing a mild and transient increase in ICP. Studies have shown that this increase is temporally related to the presence of fasciculations in the patient, but is not the result of synchronized muscular activity leading to increased venous pressure. Rather, there appears to be a complex reflex mechanism originating in the muscle spindle and ultimately resulting in an elevation of ICP. One recent study challenged the claim that SCh causes an elevation of ICP, and SCh remains the drug of choice for management of patients with elevated ICP because of its rapid onset and short duration. Although we recommended in former editions of this manual the routine use of a defasciculating agent when SCh is administered to a patient with elevated ICP, we no longer advocate this practice. There is insufficient evidence to support the use of a defasciculating agent, and it adds unnecessary complexity.

Choice of Induction Agent

When managing the patient with potential brain injury, it is important to choose an induction agent that will not adversely affect CPP. Ideally, one would like to choose an induction agent that is capable of improving or maintaining CPP and providing some cerebral protective effect. Sodium thiopental is an ultra short-acting barbiturate induction agent. Thiopental confers some cerebroprotective effect because it decreases the basal metabolic rate of oxygen utilization of the brain (CMRO₂). This effect can be likened to decreasing myocardial oxygen demand in the ischemic heart. In addition, sodium thiopental decreases CBF, thus decreasing ICP. This combination of characteristics, the decrease in ICP and the decrease in CMRO₂, make thiopental a desirable agent for use in patients with elevated ICP and a normal or high blood pressure. However, thiopental is a potent venodilator and negative inotrope. Therefore, it has a tendency to cause significant hypotension and thus reduce CPP, even in relatively hemodynamically stable patients. In the hemodynamically unstable patient, this hypotensive effect can be profound. A single episode of hypotension significantly increases mortality in acute, severe head injury. Therefore, although thiopental is a desirable agent for management of patients with elevated ICP, its hemodynamic instability relegates it to an alternative role, with etomidate being the agent of choice. When the circulating blood volume is known to be normal, and hemodynamic stability is preserved, however, thiopental remains an appropriate choice as the induction agent.

Etomidate is a short-acting imidazole derivative that has a similar profile of activity to thiopental, but without the tendency to cause hemodynamic compromise. In fact, etomidate is the most hemodynamically stable of all commonly used induction agents except ketamine (see Chapter 18). Its ability to decrease CMRO₂ and ICP in a manner analogous to that of sodium thiopental and its remarkable hemodynamic stability make it the drug of choice for patients with elevated ICP. Recently, the use of etomidate in patients with elevated ICP has been challenged on the basis of evidence from animal studies. This preliminary evidence, which is addressed in Chapter 18, does not justify avoidance of etomidate, with its excellent hemodynamic profile, in patients with elevated ICP.

Ketamine, in general, has been avoided in patients with known elevations in ICP because of the belief that it may elevate the ICP further. The evidence regarding this phenomenon is mixed, however, and is discussed in Chapter 18. In patients with elevated ICP and hypotension, ketamine's superior hemodynamic stability, on balance, argue for its use.

Approach to the Airway

RSI is the preferred method for patients with suspected elevated ICP because it provides protection against the reflex responses to laryngoscopy and rises in ICP. The presence of coma should not be interpreted as an indication to proceed without pharmacological agents or to administer only a neuromuscular blocking agent without a sedative induction drug. Although the patient may seem unresponsive, laryngoscopy and intubation will provoke the reflexes described previously, if appropriate pretreatment and induction agents are not used. Following appropriate assessment and preparation, as described in Chapter 3, the sequence in Box 28-1 is recommended for patients with elevated ICP.

Initiating Mechanical Ventilation

Mechanical ventilation in the patient with elevated ICP should be predicated on two principles: (a) optimal oxygenation, and (b) avoidance of ventilation mechanics (e.g., positive end-expiratory pressure, high peak inspiratory pressure) that would increase venous congestion in the brain.

There is no scientific basis for the use of “therapeutic” hyperventilation, with good evidence that it promotes worse outcomes rather than better. The Brain Trauma Foundation Guidelines for the Management of Severe Traumatic Brain Injury recommend that prophylactic hyperventilation be avoided and that patients with severe TBI be ventilated in such a way as to target the lower limits of normocapnia (PaCO₂ of 35–40 mm Hg). A similar approach seems prudent in patients with medically induced elevations of ICP (e.g., cerebral hemorrhage).

BOX 28-1 Rapid Sequence Intubation for Patients with Elevated Intracranial Pressure

Hyperventilation to a PaCO₂ of 30 mm Hg should be used only as a temporizing measure in patients demonstrating clinical signs of herniation (blown pupil or decerebrate posturing) and when osmotic agents, cerebrospinal fluid drainage, or both are not effective in managing an acute rise in ICP accompanied by patient deterioration. Normal initial physiological ventilation parameters are described in Chapter 37. Initial inspired fraction of oxygen (F_iO₂) should be 1.0 (100%). F_iO₂ can later be decreased according to pulse oximetry, as long as 100% oxygen saturation is maintained. Carbon dioxide tension can be followed with arterial blood gases or, preferably, continuous capnography, the first assessment of which should occur approximately 10 minutes after initiation of steady-state mechanical ventilation. To permit early and frequent neurological examinations (e.g., by a neurosurgeon to decide whether there is sufficient persisting neurological functioning to warrant an attempt at surgical evacuation of a massive subdural hematoma), long-term sedation is best accomplished by using a propofol infusion, which can be terminated as needed with prompt patient recovery. Deep sedation is desired, however, to permit effective controlled mechanical ventilation and other necessary interventions, while mitigating the stimulating effects of the tube in the trachea and eliminating any possibility of the patient coughing or bucking. An analgesic, such as fentanyl, is used to improve endotracheal tube tolerance and reduce stimulation and responsiveness.

Tips and Pearls

RSI is clearly the desired method for tracheal intubation in patients with suspected elevation of ICP. The technique allows control of various adverse effects and optimal control of ventilation after intubation. However, the use of neuromuscular blockade in patients with potential neurological deficit carries the responsibility of performing a detailed neurological evaluation on the patient before initiation of neuromuscular blockade. The patient's ability to interact with the surroundings, spontaneous motor movement, response to deep pain, response to voice, localization, pupillary reflexes, and other pertinent neurological details must be assessed carefully before administration of neuromuscular blockade. The careful recording of these findings will be invaluable for the ongoing evaluation of the patient.

If the patient's ventilatory status is severely compromised by the head injury or by concomitant injuries, positive-pressure ventilation with bag and mask may be required throughout the intubation sequence. In such circumstances, one is trading off the increased risk of aspiration against the hazard of inadequate oxygenation and rising PaCO₂ during the intubation sequence. When such a tradeoff arises, it should be resolved in favor of oxygenation over the risk of aspiration.

Evidence

Evidence-based recommendations depend on a careful analysis of the methodology used in the studies reviewed and an understanding of the outcome measure, which must be sound to make the study clinically relevant. In this light, it becomes challenging to make evidence-based recommendations regarding airway management in the patient with a brain injury. Regarding methodology, most of the studies of the effect of interventions discussed in this chapter were performed on stable patients in the operating room setting; others were performed on deeply anesthetized patients in the intensive care unit during tracheal suctioning. It is difficult to extrapolate the findings in these patient groups to critical patients undergoing emergency intubation. In addition, the timing and dosing of pharmacological interventions varied significantly, making it difficult to compare one study with another. For example, in one study lidocaine was found effective when given 3 minutes before intubation and ineffective if given at 4, 2, or 1 minute before (1). There is only one randomized double-blind interventional study identified that was performed in the ED on patients with head injury (2). This prospective double-blind study found that esmolol and lidocaine had similar efficacy in attenuating the hemodynamic response to intubation of patients with isolated head injury.

Regarding outcome, there is no study in the literature that compares airway interventions with a functional outcome measure, that is, disability or death. Rises in heart rate, blood pressure, and ICP are the commonly measured parameters comparing one technique or pharmacological intervention with the other because these affect CPP. However, there is no evidence that these are valid surrogates for more meaningful outcome measures such as disability, nor is there evidence that transient rises in any of the previously mentioned measures have any meaningful impact on morbidity or mortality. That said, there is no evidence that the interventions presented in this chapter do harm, and pending more direct evidence, it does seem intuitive that minimizing adverse changes in ICP, blood pressure, and heart rate can only contribute to maximizing good outcomes.

1. Should premedication(s) be used when RSI is performed on patients with elevated ICP? The choices of RSI premedications for patients with elevated ICP include lidocaine, fentanyl, esmolol, and a defasciculating dose of a nondepolarizing agent. Evidence for the use of lidocaine, 1.5 mg/kg IV, and fentanyl, 3 µg/kg IV, 3 minutes before induction, is discussed in Chapter 17. In summary, there is insufficient evidence that lidocaine can mitigate the RSRL, but it appears somewhat effective in limiting the intracranial response to upper airway stimulation, which is not mediated by catecholamines. Fentanyl, however, is known to blunt the reflex sympathetic response to upper airway manipulation, mitigating the extent of catecholamine release and, therefore, rise in MAP (see Chapter 17).

Esmolol has been studied as a premedication to blunt the RSRL through its beta-blocking properties. In one randomized double-blind, placebo-controlled study, esmolol, 2 to 3 mg/kg IV, provided better control of heart rate and blood pressure than either lidocaine or fentanyl (3). Of note, fixed doses of drugs were used, and the lidocaine and fentanyl were given only 2 minutes before intubation. Similar results were reported in another randomized double-blind study using 1.4 mg/kg (4) and 2 mg/kg (5). In a randomized double-blind study comparing the hemodynamic response of esmolol and lidocaine, both were found equally effective (2).

There is some controversy, but no evidence, whether the increase in ICP caused by SCh is clinically significant, and whether a defasciculating dose of a competitive neuromuscular blocking agent is capable of mitigating this response. On balance, we no longer recommend the use of a defasciculating dose (one-tenth of the paralyzing dose) of a competitive neuromuscular blocking agent such as vecuronium (0.01 mg/kg) or rocuronium (0.06 mg/kg) 3 minutes before SCh is given. Based on the best evidence available at this time, the following recommendations can be made regarding the pharmacological mitigation of exacerbations of elevated ICP during emergency intubation:

· Administer lidocaine, 1.5 mg/kg, 3 minutes prior to airway manipulation to mitigate the raise in ICP from laryngoscopy.

· In patients without compensated or decompensated shock, administer fentanyl, 3 µg/kg, 3 minutes prior to airway manipulation in order to mitigate raises in ICP from RSRL.

· When patients with elevated ICP are paralyzed with SCh, use of a defasciculating dose of nondepolarizing paralytic agent is no longer recommended.

· Esmolol is an effective agent in mitigating raises in ICP from RSRL; however, because of its potential to cause or aggravate hypotension, especially in patients with hypovolemia, it is not recommended for routine use in emergency intubation.

2. Is hyperventilation (ETCO₂, 30–35 mm Hg) recommended in the management of the TBI patient with suspected ICP? Hyperventilation can be defined as a PaCO₂<35 or an ETCO₂ <30–35: the correlation between the two measures is generally good in patients who are normotensive (6). It causes vasoconstriction and thus reduces ICP (7). Unfortunately, hyperventilation will also cause a reduction in CBF. Because CBF is reduced by almost 50% in the first days after TBI, hyperventilation poses a risk of exacerbating ischemia (8). In a randomized, controlled trial, Muizelaar et al. (9) found that patients with an initial Glasgow Coma Scale score of 4 to 5 who were hyperventilated to a PaCO₂ of 25 mm Hg during the first days after head injury had significantly worse outcomes than patients kept at a PaCO₂ of 35 mm Hg.

Despite observations that hyperventilation reverses the clinical signs of herniation, there is no evidence that hyperventilation improves outcomes (10). In an observational study, 59 adult severe TBI patients who required RSI for intubation were matched to 177 historical nonintubated controls (11,12). The study used ETCO₂ monitoring and found an association between hypocapnia and mortality, and a statistically significant association between ventilatory rate and ETCO₂. Both the lowest and final ETCO₂ readings were associated with increased mortality versus matched controls. Although the study did not specifically identify which if any of the patients demonstrated signs of herniation, it contributes to the growing body of evidence arguing against hyperventilation under any circumstance in TBI patients, and this is likely analogous to patients with medically caused elevated ICP.

Based on the best available evidence, the following recommendations can be made:

· Hyperventilation (ETCO₂ <30–35) should be carefully avoided in patients with medical intracranial catastrophe or TBI who do not demonstrate signs of increased ICP (“blown pupil” or extensor posturing).

· There is no evidence that hyperventilation improves outcome in patients with elevated ICP, and there is some evidence that it causes harm. If hyperventilation is considered, it should only be used briefly, as a temporizing measure, in the management of patients exhibiting signs of increased ICP who have failed to respond to osmotic agents.

· Intubated patients with TBI should have continuous ETCO₂monitoring in order to avoid inadvertent hypocapnia (ETCO₂ <35).

There is a clear need for a well-controlled comparative study using a meaningful outcome measure to determine which, if any, of these interventions will decrease morbidity or mortality in patients with elevated ICP undergoing emergency RSI. Pending such a study, which will likely never be done because of logistical challenges, the approach outlined in Box 28-1 seems rational. Management of the patient at risk for elevated ICP should ensure cerebral perfusion and oxygenation. When intubation is indicated, pretreatment should be provided using lidocaine 1.5 mg/kg and fentanyl 3 µg/kg. There is insufficient evidence to support the use of a defasciculating agent. Once the patient is intubated, continuous ETCO₂ should be provided to safeguard against inadvertent hypocapnia and its associated increase in mortality.

References

1. Abou-Madi MN, Keszler H, Yacoub JM. Cardiovascular reactions to laryngoscopy and tracheal intubation following small and large intravenous doses of lidocaine. Can Anaesth Soc J 1977;24:12–19.

2. Levitt M, Dresden G. The efficacy of esmolol versus lidocaine to attenuate the hemodynamic response to intubation in isolated head trauma patients. Acad Emerg Med 2001;8:19–24.

3. Helfman SM, Gold MI, DeLisser EA, et al. Which drug prevents tachycardia and hypertension associated with tracheal intubation: lidocaine, fentanyl, or esmolol? Anesth Analg 1991;72:482–486.

4. Singh H, Vichitvejpaisal P. Comparative effects of lidocaine, esmolol, and nitroglycerin in modifying the hemodynamic response to laryngoscopy and intubation. J Clin Anesth 1995;7:5–8.

5. Feng CK, Chan KH, Liu KN, et al. A comparison of lidocaine, fentanyl, and esmolol for attenuation of cardiovascular response to laryngoscopy and tracheal intubation. Acta Anaesthesiol Sin 1996;34:61–67.

6. Yosefy C, Hay E, Nasri Y, et al. End tidal carbon dioxide as a predictor of the arterial PCO₂ in the emergency department setting. Emerg Med J 2004;21:557–559.

7. Raichle M, Plum F. Hyperventilation and cerebral blood flow. Stroke 1972;3:566–575.

8. Marion D, Darby J, Yonas H. Acute regional cerebral blood flow changes caused by severe head injuries. J Neurosurg 1991;74:407–414.

9. Muizelaar J, Marmarou A, Ward J, et al. Adverse effects of prolonged hyperventilation in patients with severe head injury: a randomized clinical trial. J Neurosurg 1991;75:731–739.

10. Brain Trauma Foundation. Guidelines for the management of severe traumatic brain injury, third edition. J Neurotrauma 2007;24(Suppl 1):S1–S108.

11. Davis DP, Dunford JV, Poste JC, et al. The impact of hypoxia and hyperventilation on outcome after paramedic rapid sequence intubation of severely head-injured patients. J Trauma 2004;57:1–10.

12. Davis DP, Dunford JV, Ochs M, et al. The use of quantitative end-tidal capnometry to avoid inadvertent severe hyperventilation in patients with head injury after paramedic rapid sequence intubation. J Trauma 2004;56:808–814.