The trusted source for
healthcare information and
By Vibhu Sharma, MD
Attending Physician, Division of Pulmonary and Critical Care Medicine, John H. Stroger Hospital of Cook County; Assistant Professor of Medicine, Rush University Medical Center, Chicago
Dr. Sharma reports no financial relationships relevant to this field of study.
Ketamine is a dissociative N-methyl-D-aspartate (NMDA) antagonist with sedative and analgesic properties.1 The analgesic properties in particular stem from interaction with opioid receptors (specifically, the mu and kappa receptors). This interaction is weaker than NMDA antagonism, which is the major mechanism of action.2 The drug exerts its unique effects by “disconnecting” the thalamocortical and limbic systems, effectively dissociating the central nervous system from external stimuli (e.g., pain, sight, sound).3 Ketamine has an acceptable safety profile with a wide dose-adverse effect ratio4 and is the anesthetic of choice in settings in which there may be a paucity of resources.5 The major indications for use of ketamine in the ICU are analgosedation (a single drug for both analgesia and sedation) and induction of anesthesia around the time of endotracheal intubation. Ketamine may be used as a bolus dose alone or as a bolus dose, followed by a low-dose drip (depending on the goals of care or circumstances of use). This is a guideline for ketamine use in the appropriate clinical setting from the practicing intensivist’s perspective.
Although literature supporting its use is sparse, ketamine has been used in the adult cardiac surgery patient as an analgosedative or as an induction agent. In the operative setting, ketamine has been used as the primary anesthetic in patients with critical coronary ischemia undergoing coronary artery bypass grafting (CABG) with total IV anesthesia.6 In one study, a ketamine-based total IV anesthesia regimen (when compared with an opiate-based regimen) was not associated with a higher peak troponin leak postoperatively.6 No other safety issues were noted, and there was no evidence of harm given theoretical concerns of sympathetic augmentation. All patients in this study were hemodynamically stable and were undergoing elective CABG.
For patients with depressed left ventricular (LV) ejection fraction undergoing CABG, one study revealed that a ketamine-propofol induction combination (“ketofol”), compared to an etomidate-midazolam combination, was safe and associated with a higher cardiac index and systemic vascular resistance at one and three minutes after intubation.7 However, in a small randomized trial, Baradari et al8 found that in patients with reduced LV function and critical coronary artery disease (CAD) undergoing CABG who were randomized to similar doses of ketofol and IV etomidate, etomidate provided superior hemodynamic stability. However, when ketamine (without propofol) was compared to etomidate in an identical population of patients with reduced LV function and critical CAD undergoing CABG, no hemodynamic differences were found.9 Ketamine has been used for pain management after sternotomy in patients undergoing CABG. The authors of a randomized, placebo-controlled, double-blind trial assessed the impact of a ketamine bolus/infusion on opiate need postoperatively in this setting.10 A patient-controlled analgesia (PCA) delivery device alone was compared to PCA plus a ketamine infusion. An opioid-sparing effect was found in the ketamine infusion group without an increase in clinically significant adverse effects.
A major indication for the use of ketamine in the post-operative patient is its opioid-sparing effect. Abdominal surgery-associated ileus is an undesirable adverse effect. The opioid-sparing effect of ketamine as an adjunct to analgesia may promote the return of gastrointestinal motility. In the morbidly obese patient with either known or occult obstructive sleep apnea, there is risk for airway obstruction and hypercapneic respiratory failure related to opioid sensitivity. In these patients, the use of ketamine as an adjunct to opioids in the management of postoperative pain is supported by guidelines, although high-level supportive evidence is lacking.11 Opioid-sparing effects have not been shown to be useful among patients expected to experience mild postoperative pain (e.g., head and neck surgery). The beneficial effects of ketamine are clearer in procedures with an expectation of severe pain (e.g., thoracoabdominal procedures, spine and limb orthopedic procedures).12 The opioid-dependent or tolerant patient is a special subset in whom ketamine may be used to boost analgesia. Consensus guidelines for the use of ketamine in the postoperative setting agree that the evidence supports these opioid-sparing effects with at least a moderate level of confidence.11
In the non-cardiac surgery patient and post-operative setting, the analgesic (subanesthetic) dose of ketamine usually is a bolus not exceeding 1 mg/kg (0.3-0.5 mg/kg)13,14 or a drip at 0.5 mg/kg/hour for opioid-naïve patients and escalating to up to 1 mg/kg/hour for opioid-tolerant patients.15
Both dosing approaches define “low-dose ketamine“(LDK) for analgesia.14 These methods have been studied in the post-operative setting of spinal surgery16 and hip surgery.17 In these settings, an intraoperative bolus dose of ketamine followed by an LDK infusion for 24 hours was shown to reduce opiate requirements with no substantial increase in adverse effects. The authors of a randomized, controlled trial compared morphine PCA alone with an LDK bolus/drip (0.5 mg/kg followed by 0.1 mg/kg/hour) plus morphine PCA after major abdominal surgery and found reduction in doses of morphine use.18 Reducing opiate use in this setting may accelerate the return of bowel function. A systematic review of pooled studies comparing the use of sustained infusions of ketamine (> 24 hours) with opiate infusions in the critically ill suggested that ketamine may decrease opiate consumption in the medical ICU.19 There have been concerns about hallucinations and emergence reactions with ketamine; however, a retrospective study comparing low-dose infusion, ketamine-based sedation with non-ketamine-based regimens in the medical ICU showed equivalent rates of delirium.20 Both the pooled studies and the retrospective cohort study revealed no signal of harm or significant adverse effects with ketamine-based regimens.19,20
Ketamine is unique among induction agents in that airway reflexes are maintained and spontaneous breathing preserved while simultaneously providing deep analgesia and dissociative anesthesia.3 These properties make ketamine an ideal agent for airway management. With usual IV induction doses used for intubation, time to onset of action is 30-40 seconds. The usual endpoint for “dissociation” is a catalepsy-like state: The eyes are open with a glassy-eyed stare, nystagmus is frequent, pupillary response is intact, and involuntary movements may occur.
In scenarios of borderline hemodynamics in which avoidance of hypotension is paramount, pairing ketamine with propofol (i.e., ketofol) may be an attractive regimen. In one study, American Society of Anesthesiologists (ASA) class I and II patients undergoing induction for general anesthesia prior to elective surgery were randomized to either a ketofol bolus (ketamine 0.75 mg/kg and propofol 1.5 mg/kg) or propofol alone (2 mg/kg bolus).21 Ketofol was associated with improved hemodynamics and cardiac indices after induction and a smaller risk of a > 20% drop in systolic blood pressure after induction. The authors of a similar randomized, double-blind study of adults with normal LV function undergoing elective surgery assessed hemodynamics after the administration of ketofol (ketamine 0.5 mg/kg and propofol 1.5 mg/kg) and compared findings to IV etomidate dosed at 0.3 mg/kg.22 Ketofol was found to be safe and associated with superior hemodynamic stability. Two small case series in the critically ill and the elderly undergoing induction prior to surgery seem to add to the safety profile of the combination ketofol prior to endotracheal intubation in patients with normal LV function.23,24
Other studies have compared ketofol with propofol alone prior to insertion of laryngeal mask airways (LMAs) in the setting of elective surgery. The authors of a study comprised of elderly (age > 65 years, mean age = 72 years) patients (ASA class I and II) undergoing LMA insertion prior to surgery randomized patients to propofol alone or ketofol.25 Ketofol was associated with higher blood pressure right after induction and at five minutes after induction compared with propofol alone. In a similar study, researchers randomized adult patients to either ketamine, saline, or fentanyl prior to induction with propofol.26 Ketamine was associated with a significantly higher systolic blood pressure compared with the other agents. Insertion conditions subjectively assessed by anesthesiologists with an ordinal scale assessing gagging and laryngospasm among others were no different in either group (except saline, where conditions were notably poorer). No safety concerns were noted in either study. These studies provide a framework for using ketofol in the critically ill and in the aggregate. It can be inferred that ketofol is not associated with laryngospasm, may be hemodynamically neutral in the absence of LV dysfunction, likely is safe in the elderly, and may be better tolerated hemodynamically than propofol alone in the setting of airway management.
The authors of a large emergency department (ED) study analyzed 469 patients randomized to rapid sequence intubation (RSI) with either etomidate (0.3 mg/kg IV bolus) or ketamine (2 mg/kg IV bolus), followed by succinylcholine.27 The primary aim was to assess the incidence of adrenal insufficiency in each group. This trial is instructive with respect to the broad inclusion criteria. Patients with a history of stroke, CAD, or chronic heart failure were included. A large proportion of patients were intubated secondary to trauma without regard to whether traumatic brain injury (TBI) was present. Others had suffered an acute stroke or were in cardiogenic shock. The only exclusion criteria were cardiac arrest, pregnancy, or relevant allergies to drugs. The heart rate in the ketamine group at enrollment ranged from 65-128 beats/minute and systolic blood pressure 96-160 mmHg. The proportion of cardiac arrests during intubation were equivalent in each group. In addition to confirming the lower odds of adrenal insufficiency compared to etomidate, this randomized trial confirms the safe use of ketamine in a wide selection of critically ill patients.
Ballow et al28 developed a protocol for intubation of critically injured patients in the ED that included ketamine administered as a bolus dose of 1-3 mg/kg followed by either succinylcholine or rocuronium. Ketamine replaced etomidate/midazolam/fentanyl as the induction agent. Ketamine was administered regardless of the patient’s age or whether TBI had occurred. These investigators noted an improvement of time to intubation and no safety concerns.
Weingart et al discussed the use of delayed sequence intubation to expand on the role of ketamine in the setting of airway management.29 The authors described a prospective convenience cohort of 62 adult ED patients, most presenting with a primary pulmonary condition (pneumonia, asthma, lung injury, pulmonary edema, COPD) as the sole indication for intubation. Other indications included cardiogenic shock and trauma. Each patient in this cohort entered the study with difficulty preoxygenating with either a mask or noninvasive positive pressure ventilation (NIPPV) due to delirium or agitation; seven out of 62 patients were located in the critical care unit. Ketamine was administered with an initial bolus dose of 1 mg/kg followed by additional 0.5 mg/kg boluses as needed until the patient was completely dissociated. The mean total dose of ketamine for the entire cohort was 1.4 mg/kg. Thereafter, preoxygenation was accomplished with either a mask or NIPPV at 5-15 cm H20 with no support provided given spontaneous respirations. Muscle relaxants were administered for intubation. No safety issues were identified in this study. All patients improved their oxygen saturations, including those with critical desaturation (93%) prior to DSI.
Ketamine offers a plausible pharmacologic mechanism for use in the setting of severe alcohol withdrawal in that both ketamine and alcohol antagonize the NMDA receptor.30 The authors of a retrospective study (pre- and post-guideline methodology) assessed the effect of LDK infusion added (post-guideline) to usual therapy with benzodiazepines titrated to symptoms (pre-guideline) on requirements for benzodiazepines in patients admitted with delirium tremens (DT).31 LDK was initiated as soon as the diagnosis of DT was recognized. The LDK infusion dose was 0.15-0.3 mg/kg/hour (median, 0.24 mg/kg/hour), and a bolus (0.3 mg/kg) was provided at the outset as needed. Dexmedetomidine, propofol, benzodiazepines, or phenobarbital were used as indicated pre- and post-guideline. LDK infusion continued for a median of approximately two days (range, 35-71 hours). Phenobarbital and benzodiazepine doses administered were converted into a “diazepam-equivalent” (DE) dose for the purposes of the study. LDK infusion started at the time of DT diagnosis significantly reduced the rate of endotracheal intubation (29% intubated with LDK vs. 76% intubated pre-guideline), reduced ICU length of stay (5.7 days with LDK vs. 11.2 days pre-guideline), and reduced gamma-amino butyric acid (GABA) agonist administration (834 mg total DE with LDK infusion vs. 3,016 mg total DE pre-guideline). The auhtors of another retrospective study assessing benzodiazepine-refractory alcohol withdrawal also noted a decrease in benzodiazepine use and quick symptom control after initiation of a ketamine infusion.32 No adverse neurological effects or hemodynamic perturbations of concern were noted even though infusion rates were higher (median = 0.75 mg/kg/hour) compared to the study referenced above.31 Randomized trials are needed to confirm these findings.
Super refractory status epilepticus (SRSE) is defined as status epilepticus (SE) continuing after 24 hours of anesthetic infusion or recurrence after discontinuation of anesthetic infusion.33 Typically, SE is treated with a combination of benzodiazepines and specific antiepileptics, quickly escalating to anesthetic infusions (most commonly propofol). Prolonged SE and SRSE are associated with a downregulation of GABA-A receptors and an upregulation of NMDA receptors, nicely positioning ketamine as a therapeutic adjunct, especially given concerns for adverse effects of prolonged propofol infusions, such as hyperlipidemia and propofol infusion syndrome.33 There are no randomized trials comparing ketamine to other agents in SRSE. Usually, consideration is given for use late in the disease course prior to stepping up to inhalational general anesthesia. Sabharwal et al studied patients, most of whom were on both propofol and ketamie infusions, in a neurocritical care setting.33 Usual dosing in this setting is an infusion with a dose range of 1.5 mg-10.5 mg/kg/hour added to a propofol infusion (1.5-8 mg/kg/hour). A multinational case audit of the treatment of SE revealed that ketamine was considered only for the most severe cases and then late in the course of the disease.34 More recent experience with ketamine and a possibility for a neuroprotective effect make the case for early administration (duration of SE < 3 days) rather than late rescue.35,36 An Italian group reported their experience in children on a protocol that used IV ketamine for refractory SE upfront added to other antiepileptics. They found that in instances in which ketamine was used as the sole anesthetic agent for refractory SE, there was the added benefit of avoiding endotracheal intubation.37 This experience has not been replicated in adults but is intriguing.
Small randomized trials and case reports support the role of ketamine in patients with status asthmaticus consistent with its physiologic effects on smooth muscle in the lung.38-40 When used in this setting, dosing typically is similar to the LDK infusion used for analgosedation in the postoperative setting (0.5 mg/kg/hour) and may be titrated up to 2 mg/kg/hour, with duration of infusion ranging in various studies from one hour to five days.39
There is limited experience with ketamine in the setting of pulmonary hypertension (PH). There is a clear risk for hemodynamic perturbations in the critically ill with PH regardless of what drug is used for RSI. Both etomidate and ketamine seem appropriate for RSI in this setting.41 Although no conclusions can be reached for the safety of ketamine in adults with PH, given the near nonexistence of literature in this setting, ketamine has been used safely for surgery and procedural sedation in children with pulmonary hypertension.42
As with any drug, it is important to be familiar with adverse effects that can be expected with its use in the critical care setting. Ketamine appears to have a wide margin of safety. In one survey of physicians practicing in the developing world, where ketamine was used frequently in nonoperating room settings and without any monitoring,17 complications potentially related to ketamine were reported in more than 12,000 administrations (e.g., apnea [n = 10], laryngospasm [n = 6]).5 There is a paucity of systematic reviews of ketamine use and safety in the critical care setting, but expectations can be inferred from the emergency medicine and anesthesia literature. A review of the adverse effects of ketamine when used for procedural sedation in the ED for adults (dose of up to 2.5 mg/kg followed by smaller supplemental doses as needed or 2 mg/kg intramuscular) revealed that it was generally safe, with one cardiorespiratory arrest reported in a single debilitated adult out of 70,000 reported patient contacts.43 Ketamine-associated adverse effects may include purposeless movements, nystagmus, hypersalivation, vomiting, emergence reactions, laryngospasm, a transient increase in heart rate and blood pressure, and respiratory depression and apnea. Arguably, these adverse effects may not apply to patients who are critically ill, but are instructive regarding what to expect.
There has been debate about the use of ketamine in the patient with elevated intracranial pressure (ICP). Experimental animal models of cerebral trauma have shown that there is no increase in cerebral edema with escalating doses of ketamine.44 The authors of one systematic review included 10 randomized trials (none blinded) that compared ketamine with opiates (sufentanil, remifentanil, or fentanyl) or etomidate for intubation in the setting of severe TBI in the ED setting or in the ICU for patients with subarachnoid hemorrhage (SAH) or craniotomy for tumor or hydrocephalus.45 Seven of these studies were conducted in an ICU; the rest were conducted in the operating room where midazolam or propofol often were administered concurrently. The authors identified no increases in ICP or changes in cerebral perfusion pressure (CPP) of consequence but stated that a lack of large, high-quality, adequately powered, randomized trials limited the strength of their conclusions.
In another study, investigators compared fentanyl to ketamine for sedation in the setting of SAH and TBI in the neurocritical care setting and found no elevations of ICP or changes in CPP of note.46 Patients receiving ketamine showed a trend toward a lower requirement for vasopressor support. The authors concluded (with respect to using ketamine in the neurosurgical critical care setting) that “its use in neurosurgical patients should not be discouraged on the basis of ICP-related concerns.”46 In a systematic review of randomized, controlled trials, researchers reached similar conclusions.47 Most clinical studies recruiting patients in the ED have not specifically excluded patients with head trauma. Until there is more clarity on this topic in the critical care setting, it may be prudent to use an alternative induction agent in patients with uncal/transtentorial herniation or impending herniation and in those with ventricular obstruction/obstructive hydrocephalus. Ketamine seems safe as an induction agent for intubation in the patient with SAH, among those patients with intracranial mass lesions and cerebral edema, as well as patients with TBI.
Financial Disclosure: Critical Care Alert’s Physician Editor Betty Tran, MD, MSc, Nurse Planner Jane Guttendorf, DNP, RN, CRNP, ACNP-BC, CCRN, Peer Reviewer William Thompson, MD, Executive Editor Leslie Coplin, Editor Jonathan Springston, Accreditations Manager Amy M. Johnson, MSN, RN, CPN, and Editorial Group Manager Terrey L. Hatcher report no financial relationships relevant to this field of study.