Explanation of the bleeding edge series here.

Introduction with a case

Once upon a time at Genius General Hospital, I arrived a bit early for a shift.  The outgoing attending was happy to see me:  “Hi Josh, you got here just in time to intubate the patient in Bed 8.  He was supposed to get an upper endoscopy with ketamine sedation, but he couldn't be sedated.”

The patient in Bed 8 was a 40-year-old man with a history of alcoholism, currently presenting with an upper GI bleed.  He had received 130 mg of ketamine (2 mg/kg), which caused him to become partially dissociated and mildly agitated.  Airway equipment was at the bedside and preparations were underway to perform intubation.

Instead of proceeding with intubation, I decided to attempt giving more ketamine.  We gradually up-titrated the ketamine dose in increments of 70-100 mg.  He ultimately required a total of 500 mg IV ketamine in order to reach full dissociation (~8 mg/kg).  Endoscopy was performed without problems and his recovery was uneventful.

The clinical phenomenon of ketamine tolerance

This is one of about a half dozen patients I've encountered with ketamine tolerance.  These patients do not fully dissociate at 2 mg/kg, a dose which should be adequate.  Tolerance can be overcome by using higher doses of ketamine (e.g. 400-600 mg IV in divided doses).

Ketamine tolerance seems to be limited to patients with a history of substance or alcohol abuse.  Unfortunately I haven't been systematic about obtaining rigorous histories in these patients, so I'm not sure.

This phenomenon isn't widely reported in the literature or well known among clinicians.  Most guidelines and review articles indicate that a dose of 2 mg/kg will reliably cause dissociation (e.g. Green 2011).

The ketamine-tolerant patient presents a quandary to clinicians who aren't familiar with this phenomenon.  The first time I encountered this, I was baffled and aborted the procedure after giving 200 mg ketamine.  Eventually I realized that the drug isn't “failing” to work, but rather we are failing to administer a sufficiently high dose.

Evidentiary support for ketamine tolerance

Two reports from the 1980s documented the development of tolerance to ketamine anesthesia among patients treated with repeated doses (Khan 1988, MacLennan 1982).  These are both freely available online and reproduced here in full:


More recently, several reports have documented tolerance to ketamine's psychoactive effects (Bonnet 2015).  This is a well-known phenomenon in the context of illicit ketamine abuse.  Thus, the ability to develop tolerance to ketamine is established in the literature.

Ketamine tolerance has been proven to exist in rats and monkeys.  Tolerance evolves rapidly in both species (figure below).  Among rats, cross-tolerance exists between ketamine and fentanyl (but not morphine).  This raises the possibility that cross-tolerance in humans could exist between ketamine and some synthetic opioids.

Ketamine causes dissociation by inhibiting NMDA receptors.  Other drugs which inhibit NMDA receptors include ethanol, gabapentin, nitrous oxide, dextromethorphan, methadone, and phencyclidine.  It might be predicted that patients who use or abuse such substances would develop cross-tolerance with ketamine.  Cross-tolerance between ketamine and either phencyclidine or ethanol has been demonstrated in animal models (Fidecka 1989; Wegner 1982; Rocha 1995).

The safety of high-dose ketamine

There is relatively little evidence in the literature regarding ketamine tolerance.  However, there is substantial evidence that high doses of ketamine are safe.  For example, intramuscular doses of 500 mg have been utilized for the management of agitated patients.  The FDA-approved ketamine package insert recommends a dose of 9-13 mg/kg for intramuscular induction of anesthesia (1).  The literature contains examples of inadvertent administration of massive doses of intravenous ketamine with minimal adverse effects (aside from prolonged dissociation; Green 1999).

The safety of ketamine is based upon its mechanism of action as a dissociative anesthetic.  It's impossible to “over-dissociate” a patient: once they are dissociated, additional drug can't dissociate them further.  This creates a ceiling effect on the neurologic action of ketamine.  There may be some concern about hemodynamic effects of very large doses, because ketamine can cause catecholamine release (increasing the Bp) and also has some negative inotropic properties (potentially dropping the Bp).  However, with gradual dose titration and hemodynamic monitoring this should be safe.

Ketamine to facilitate procedures in the “difficult to sedate” patient

We are sometimes called upon to perform procedural sedation for non-intubated patients with a history of being “difficult to sedate.”  A common scenario is a patient with alcoholism who requires an endoscopy or bronchoscopy.  Such patients often respond poorly to a traditional sedation regiment for endoscopy (midazolam and fentanyl), due to cross-tolerance between alcohol and benzodiazepines.

Ketamine monotherapy is a uniquely useful drug for this situation.  Even in a patient with ketamine tolerance, eventually a high enough dose should achieve full dissociation.  As discussed above there is little risk of “overdosing” the patient with ketamine.  This is in distinct contrast to other commonly used agents (e.g. midazolam, fentanyl, propofol), where escalating doses do increase the risk of respiratory complications.

Some pitfalls to using ketamine for procedural sedation

In fairness, the following points should be mentioned:

  1. Bolusing ketamine may cause short and self-limited apnea (~30 seconds). This may be avoided by giving each dose gradually (e.g. 100 mg over a minute).  As with any procedural sedation, the provider should be ready to provide a jaw thrust and a few breaths via a bag-valve mask if needed.
  2. According to textbooks, ketamine causes retention of airway muscle tone. However, I have seen patients with obstructive sleep apnea who develop airway obstruction following ketamine.  This may be managed with basic maneuvers including airway repositioning and a nasopharyngeal airway (note that ketamine doesn't suppress vomiting, so patients who are fully dissociated can still vomit in response to placement of an oropharyngeal airway).
  3. Some individuals who abuse high-dose ketamine could conceivably develop a level of tolerance so great that they wouldn't dissociate even in response to large doses of ketamine (e.g. 10 mg/kg).  I'm not aware of any evidence that this exists, but it might (2).

Thus, giving ketamine isn't entirely foolproof.  However, when titrated with attention to airway and vital signs, it is very safe.

Ketamine tolerance & rapid-sequence intubation

When performing rapid sequence intubation, we generally give the patient a standard dose of sedative (e.g. 2 mg/kg ketamine) with the assumption that this will cause dissociation.  Unfortunately, if this assumption is wrong, then it is possible for a patient to be paralyzed without adequate sedation.  This is problematic regardless of what paralytic is used, but it is especially undesirable with rocuronium (because it is possible for the patient to have an extended period of paralysis with awareness).

There seem to be two main risk factors for awareness during rapid-sequence intubation using ketamine:

  1. History of substance abuse or alcoholism.
  2. Patient has intact mental status prior to intubation (e.g. patients who aren't that severely ill, who are being electively intubated to facilitate a procedure). Most critically ill patients are delirious prior to intubation, which provides some protection from awareness.

For patients with these risk factors, an alternative strategy is needed.  This scenario can generally be managed nicely by a “delayed-sequence intubation“-style approach.  Ketamine may be gradually up-titrated until the patient is dissociated.  After the patient is dissociated, intubation may be performed with confidence that the patient is adequately sedated.  A similar dose of ketamine may be repeated after 30 minutes to maintain sedation if rocuronium was used.

  1. Intramuscular ketamine is 93% bioavailable, so presumably doses of 9-13 mg/kg IM would be equivalent to administration of a similar cumulative dose of IV ketamine in divided doses over several minutes.
  2. The recklessness of substance abusers and plasticity of the human brain almost certainly exceed the courage of physicians. Thus, someone is probably out there somewhere abusing ketamine at doses greater than any physician would have the courage to use.  This level of tolerance (the “super-tolerant” patient), if it exists, is probably exceedingly rare.
Josh Farkas
Social Me

Share this:

Would love your thoughts, please comment.x