The first step in pharmacologic therapy advocated in the most recent ACCM PAD guidelines is pain control, for pain in and of itself—whether from a fracture, incision, or even suctioning of an endotracheal tube (ETT)—can be anxiety provoking (2).
NONSTEROIDAL ANTI-INFLAMMATORY DRUGS
Nonsteroidal anti-inflammatory drugs (NSAIDs) do not have sedative properties, but to the extent that they decrease pain, they do decrease pain-associated anxiety. NSAIDs are often contraindicated in ICU patients because of their side effects. However, when cyclo-oxygenase (COX)-2 inhibitors, which have fewer side effects, are coupled with gabapentin or its precursor, pregabalin, they have analgesic and sedative properties if given preoperatively per os to patients who are anticipated to be admitted postoperatively to the ICU. In one study, a combination of 400 mg of celecoxib and 150 mg of pregabalin improved patients’ sedation levels by approximately 33% for up to 24 hours postoperatively (27).
Opioids
Because pain or discomfort from ETT suctioning, nasogastric tubes, or indwelling urinary bladder catheters is a frequent, confounding factor for anxious patients, analgesics are often administered, usually via continuous intravenous (IV) infusions. Importantly, opioids have not only analgesic, but also anxiolytic, properties (28) and therefore are commonly administered. In one multicenter study of endotracheally intubated, mechanically ventilated patients, opioids were administered (90% of patients) more frequently than were sedative drugs (72% of patients) or nonopioid analgesics (33% of patients) (29).
Morphine has anxiolytic properties, but is not nearly as effective as newer opioids and, because of the buildup of active metabolites in patients with renal insufficiency, it is not recommended for patients in the ICU. Fentanyl and remifentanil are over 90% effective in providing adequate sedation for endotracheally intubated and mechanically ventilated patients in the ICU (30). Because of equal efficacy and differences in cost, continuous infusions of fentanyl are recommended in most patients, except those with significant renal/hepatic impairment.
Benzodiazepines
For many years, benzodiazepines, which potentiate the effects of gamma-aminobutyric acid via the benzodiazepine receptor and suppress central nervous system (CNS) activity, were the most commonly administered drug to provide sedation in the ICU. However, a meta-analysis of reports studying the use of benzodiazepines compared to propofol or dexmedetomidine in mechanically ventilated patients demonstrated prolongation of the duration of mechanical ventilation and ICU length of stay in those patients receiving benzodiazepines (31). The current ACCM PAD guidelines recommend avoiding their use unless alcohol or benzodiazepine dependence prior to hospitalization is suspected (2). For acute alcohol withdrawal, a benzodiazepine is certainly appropriate, but once the acute episode is treated, an SSRI (selective serotonin reuptake inhibitor) may be a better choice for long-term treatment (32). In a recent study of more than 1,000 patients who were critically ill, continuous benzodiazepine administration was shown to increase the risk of developing delirium (32). Therefore, if benzodiazepines are administered, intermittent IV bolus therapy may be more appropriate.
Propofol
Propofol (di-isopropylphenol) is a highly lipophilic compound formulated in an isotonic oil in water emulsion (Intralipid) that is unrelated to other sedative/anesthetic agents. Because it is formulated in lipid emulsion, side effects include hypertriglyceridemia and bacterial contamination of infusions. The addition of the preservatives ethylenediaminetetra-acetic acid (EDTA) or bisulfite decreases the incidence of bacterial overgrowth. A rare side effect is the propofol infusion syndrome (33), presenting as metabolic acidosis, and ventricular fibrillation in children and in young adults with neurologic injury who are receiving greater than 100 μg/kg/min of propofol for greater than 12 to 24 hours. Propofol is now the most commonly used IV anesthesia induction agent, and is increasingly being used in moderate sedation (endoscopy suite) protocols. Because of its rapid onset and offset, few residual aftereffects, and low side-effect profile, it is often used for short-term sedation in the ICU. And, as the cost of the product has decreased, it is more commonly administered for long-term ICU sedation. Propofol has no analgesic properties, so for patients with pain, an analgesic drug should be co-administered. The combination of propofol and haloperidol, compared to propofol and midazolam, for treating agitation in ICU patients was found to maintain stable hemodynamics while minimizing respiratory depression. The dose of propofol necessary to calm the patients was found to be less when combined with haloperidol than would have otherwise been necessary (34). Propofol has also been used to treat status epilepticus (35) and to induce sleep in the ICU (36).
Dexmedetomidine
Alpha-2 (α2)-agonists, such as methyldopa and clonidine, have long been known to have sedative properties; in fact, clonidine is administered epidurally for its antinociceptive effects in the spinal cord. Dexmedetomidine is an α2-agonist that acts by binding to α2-receptors in the locus ceruleus with a high α2/α1 ratio of approximately 1,620:1, approximately seven times more avidly than clonidine. Binding to the α2-receptor releases norepinephrine and decreases sympathetic activity; the net effect is sedation, analgesia, and amnesia. Dexmedetomidine is unique compared to the other anxiolytic drugs because patients are not only calm but appear to be “sleeping” (37). Many clinicians prefer dexmedetomidine to opioids or benzodiazepines because the former agent is not associated with respiratory depression; dexmedetomidine is preferred to propofol because there is less hemodynamic compromise. Mirski et al. (38) demonstrated in a small study of only 30 patients that an infusion of dexmedetomidine and fentanyl attenuated anxiety and agitation better than did an infusion of propofol and fentanyl. However, because it is a central α-agonist, hypotension and bradycardia can, and do, occur. Fortunately, low-dose dexmedetomidine (6 μg/kg/hr for 10 minutes followed by an infusion of 0.2 μg/kg/hr) is as effective as higher doses (0.6 μg/kg/hr), with fewer side effects (39).
Butyrophenones
Butyrophenones are neuroleptic drugs also known as antipsychotic drugs or major tranquilizers. They induce apathy, a state of mental detachment, in patients with psychoses or delirium. By inhibiting dopamine-mediated neurotransmissions in the CNS, they decrease the frequency of hallucinations, delusions, and other abnormal thoughts. Patients become so detached from their environment that they develop a characteristic flat affect. Butyrophenones are active in the chemoreceptor trigger zone in the brainstem and thus are effective antiemetics; they are also used to treat hiccups and are used as synergistic anxiolytic drugs when used with benzodiazepines.
Of the butyrophenones, haloperidol is the drug used most often to treat delirium in the ICU; the agent has not, however, been shown to prevent delirium (40). Haloperidol has a wide therapeutic margin, but has important side effects, including hypotension, extrapyramidal symptoms, anticholinergic effects (tachycardia, urinary retention, ileus), neuroleptic malignant syndrome, and seizures; these side effects are, fortunately, rare. Hypotension following a dose of haloperidol is almost always seen in patients who are hypovolemic. Extrapyramidal symptoms are more often seen in younger patients and in patients with depleted dopamine stores, e.g., patients with Parkinson disease.
The initial dose of haloperidol is usually 0.5 to 2 mg administered parenterally; depending on the patient’s size, age, and degree of agitation or delirium, 5 mg may be given. Haloperidol has a slow onset, so peak effects may not be seen for 15 to 30 minutes; repeat doses then should be administered at 30- to 60-minute intervals. Recurrence of agitation, or an increase in delirium, is an indication for repeat doses, which may be increased if the initial dose was inadequate. Tardive dyskinesia or neuroleptic malignant syndrome can occur even during the short duration of therapy used in the ICU. Haloperidol, because of its anticholinergic effects, may prolong the QT interval on the EKG in a dose-dependent fashion, resulting in dysrhythmias, up to and including torsades de pointes; thus, patients receiving haloperidol should have continuous electrocardiographic monitoring.
A recent retrospective study of 989 mechanically ventilated ICU patients found that those who received haloperidol had significantly lower mortality than those who did not (41). Although not an indication for the use of haloperidol in all ICU patients, the results should be reassuring to those who have concerns about its use.
Ketamine
Ketamine is a rapid-acting, phencyclidine derivative, general anesthetic administered parenterally to induce anesthesia. Ketamine induces “dissociative anesthesia” because it interrupts association pathways of the brain before blocking sensory pathways—patients may perceive pain, but it does not bother them. However, because it is a phencyclidine derivative, 10% to 20% of adult patients may have psychological sequelae including hallucinations, especially at high doses.
Ketamine is often used as a general anesthetic in patients with hemodynamic instability because it raises CO, HR, and arterial and venous pressures, and it maintains pharyngeal and laryngeal reflexes without suppressing respiration. Ketamine is also a bronchodilator and has been advocated as the anesthetic agent of choice in patients with reactive airways disease. Ketamine was quite commonly used 20 to 30 years ago and, because of the increasing incidence of reactive airways disease, there is renewed interest in the use of ketamine for sedation of patients with lung disease. A 1-mg/kg bolus of ketamine can be administered, followed by an infusion of 0.5 mg/kg/hr, titrated up to 4.5 mg/kg/hr; many administer a benzodiazepine to reduce the frequency of psychological sequelae. In one study of 30 patients receiving primarily ketamine for sedation, the incidence of side effects was 13% (4/30). Two of the patients developed tachydysrythmias, and two became agitated, though secondary to the ketamine. The incidence of side effects was similar to what the investigators had seen with other sedative agents (42). Ketamine is contraindicated in patients with cardiac ischemia or elevated intracranial pressure (ICP). However, the latter prohibition is gradually being displaced; one recent review of the literature of studies of patients with traumatic brain injury did not identify a correlation between ketamine use and ICP (43).
Other Agents
Several anesthetic agents have been tried to sedate patients in the ICU, with unanticipated results. When nitrous oxide was used, anemia developed and led to the realization that nitrous oxide interfered with vitamin B12 metabolism. Similarly, when etomidate was used for sedation, patients developed adrenocortical insufficiency because we now know that etomidate interferes with cortisol metabolism.
NEUROMUSCULAR BLOCKADE
Despite what should be effective doses of anxiolytic drugs, some patients remain delirious and agitated, and a further increase in the dose of anxiolytic drugs is proscribed because of side effects. Such patients, along with those with closed-head injuries, tetanus, and ALI, may require other therapeutic modalities (Table 149.2). If the patient has an ETT in place, is mechanically ventilated, and is receiving adequate sedation, chemical paralysis with an NMBA is an option.
Patients with ARDS are often difficult to ventilate and are commonly agitated, hemodynamically unstable, and have a decreased that may be life-threatening. Additional sedative drugs will only worsen the hemodynamic status; NMBAs may be the only (life-saving) alternative that have been shown to improve gas exchange (44).
Three studies, conducted by the same investigators in French ICUs, of more than 400 patients with ARDS who were assigned at random to receive either an NMBA (cisatracurium) or placebo for 48 hours demonstrated an improvement in outcome (oxygenation and mortality) (44–46). The incidence of muscle weakness was not increased significantly in the group receiving an NMBA (46). A more recent study in China of 48 patients with ARDS assigned at random to receive vecuronium or control demonstrated similar results. The group that received vecuronium had better outcomes, both in terms of mortality and in morbidity (47). The benefit to NMBAs in ARDS most likely derives from decreased ventilator asynchrony which, in turn, may decrease airway pressures. The largest trial reported to date and a systematic review of the literature suggest that reduction in airway pressures, especially decreased plateau airway pressures, can prevent ventilator-associated lung injury and decrease ARDS mortality (48,49). The available evidence supports the use of an NMBA in adults with ARDS whenever plateau airway pressures are greater than 30 to 35 cm H2O.
TABLE 149.2 Indications for the Management of Patients with Neuromuscular Blocking Agents |