Propofol Withdrawal Syndrome in an Adult Patient With Thermal Injury

Michael J. Cawley, PharmD; Timothy M. Guse, PharmD; Anjali Laroia, PharmD; Linwood R. Haith, MD; Bruce H. Ackerman, PharmD


Pharmacotherapy. 2003;23(7) 

In This Article


Propofol was first administered to induce general anesthesia; however, recently published guidelines for administering sedatives in critically ill patients recommend that propofol be given for short-term (< 3 days) sedation, except in neurosurgical patients, due to the benefits of reducing elevated intracranial pressure.[10] This change in recommendation has been controversial, and its broad acceptance may be associated with an increase in the high acquisition cost of the drug. Furthermore, long-term (> 7 days) administration is also associated with an adequate level of sedation and has been associated with a shorter time to extubation in mechanically ventilated patients than in those managed with midazolam.[1]

Some advantages of propofol are its favorable pharmacodynamic and pharmacokinetic properties, such as high lipophilicity, which allows rapid penetration into the CNS, rapid onset of and emergence from sedation, short pharmacologic serum half-life, absence of pharmacologically active metabolites, and rapid metabolic clearance. Increased propofol administration was recently reported by 35% of intensive care physicians managing critically ill patients.[11]

Administration of long-term infusions of propofol is controversial and is no longer supported by the manufacturer (E. Bermudez, AstraZeneca Pharmaceuticals, written communication, 2000). Case reports have identified long-term and ultra-long-term propofol infusions ranging from 7-63 days with and without evidence of toxic effects due to prolonged administration.[12,13] Adverse effects associated with both long-term and ultra-long-term infusions have been identified, such as cardiovascular depression, necrotizing pancreatitis, emerging drug tolerance, and respiratory depression.[13,14,15,16] Many clinicians propose that propofol offers an advantage of easier dosage titration for assessment of daily neurologic function when compared with benzodiazepine infusions. However, several episodes of pancreatitis and withdrawal reactions associated with prolonged propofol infusions warrant concern among staff in intensive care units without protocols for short- and long-term administration.

Benzodiazepines exert their sedative and anxiolytic effects through their facilitation of the inhibitory actions of GABAA receptors in the CNS. Withdrawal symptoms can be observed after discontinuation of long-term infusions of benzodiazepines; these symptoms typically are manifested as tremors, hyperthermia, diaphoresis, hypertension, tachycardia, and seizures.[17] Symptom onset is unpredictable based on the variance in pharmacokinetics of the drug and duration of patient exposure. Lorazepam withdrawal symptoms may require several days to manifest after drug discontinuation. Withdrawal symptoms after diazepam administration, due to the drug's longer pharmacologic half-life and many active metabolites, may occur within 7 days of drug discontinuation.

Benzodiazepines and their active metabolites can prolong the pharmacologic action of the specific agent up to 200 hours after drug discontinuation; onset of withdrawal symptoms can occur for up to 3 weeks after benzodiazepine discontinuation.[18] Propofol, like the benzodiazepines, also produces anesthetic and sedative effects through modulation of GABAA receptors, and this common modulation pathway with the benzodiazepines may predispose patients to propofol withdrawal symptoms.

Table 1 summarizes published case reports of patients experiencing propofol withdrawal syndrome. The patients differ in age, admitting diagnosis, duration of propofol administration, withdrawal symptoms, and outcome. It is difficult to provide conclusive recommendations for prevention of propofol dependence and withdrawal symptoms given the limited published data. Much of the current experience with dependence and withdrawal is available only from the manufacturer.[19]

One study investigated the frequency of acute withdrawal in mechanically ventilated intensive care patients receiving analgesic and sedative drugs.[20] The authors found that patients experiencing withdrawal symptoms were more likely to have received neuromuscular blocking agents or propofol during their intensive care unit stay (18.6 ± 5.0 days) than those who did not experience withdrawal symptoms (6.6 ± 1.9 days, p<0.05).[20] Although these data further implicate the association between prolonged propofol infusions and withdrawal syndrome, the investigators were unable to identify a risk for prolonged propofol infusions exceeding a specific number of days as the definitive cause for precipitous withdrawal symptoms. More intense symptoms may reflect prolonged administration of high dosages of opioids and benzodiazepines rather than the withdrawal of propofol infusions exclusively.

The limited and inconclusive information concerning propofol withdrawal syndrome provides little direct guidance to the administration of propofol and the risk of sudden and profound drug withdrawal. However, we are convinced that the sudden onset of symptoms experienced by our patient after several attempts at drug discontinuation demonstrate a clear case for propofol withdrawal syndrome. Withdrawal symptoms were observed after two separate attempts to reduce the infusion rate and abated only on restarting the propofol infusion and increasing the infusion rate, thus representing a true rechallenge for the adverse effects that occurred. Also, unlike previous reports, our patient experienced other classic symptoms, such as abrupt agitation, tremors, tachycardia, tachypnea, and hyperpyrexia.

A confounding factor that prevents definitive implication of propofol as a cause of withdrawal symptoms is our patient's history of bipolar disease. The patient may have had periods of breakthrough manic episodes during propofol withdrawal. Although the successful administration of propofol for the treatment of refractory delirious mania is documented in the literature,[21] it may be difficult to identify whether the restart of the propofol infusion successfully treated propofol withdrawal or an acute manic episode. Unfortunately, no data we know of support the administration of propofol infusion for either the treatment of bipolar disorder in intensive care patients or for assistance with ventilatory weaning in patients with this diagnosis.

The pharmacokinetic and pharmacodynamic properties of drugs may be unpredictably altered as a result of thermal injury. Thermal injury not only affects the perfusion of elimination organs, but also causes release of cytokines that augment the expression of microsomal cytochrome P450 (CYP) 3A4 enzymes. Thus, the result may be functional reduction or induction of specific enzymes involved in specific drug metabolism.[22] The acute stress of thermal injury augments protein synthesis by increasing the synthesis of acute-phase reactant proteins, such as 1-acid glycoprotein (orosomucoid), a protein associated with augmenting the free fraction of alkaline-based pharmaceutical agents.[23] Propofol is metabolized mainly by the glucuronidation pathway; however, the metabolism of propofol has not been characterized in patients with thermal injury.

Patients with thermal injury acutely experience decreased levels of plasma proteins due to fluid shifts, capillary leaking of plasma proteins, and changes in plasma oncotic pressure. Since propofol has exhibited significant binding characteristics to serum albumin ranging from 96-99%, one may hypothesize that the free fraction of propofol blood concentrations or, alternatively, the increased intrinsic clearance of propofol in our patient, was higher.[24,25] The result may be increased distribution to deep tissue compartments and increased physical dependence. Possible altered propofol disposition further supports the observation that our patient's withdrawal was analogous to discontinuation of high-dose, long-term administration of other drugs modulating the action of GABAA receptors.

Limitations associated with this case report include the lack of reporting of serum triglyceride levels as additional monitoring for pancreatic toxicity. Patients with severe hypertriglyceridemia may have a normal serum amylase level (serum lipids interfere with the amylase assay), resulting in slowly progressing necrotizing pancreatitis.[13] This potential laboratory assay interference was unknown to the medical team at the time; therefore, amylase and lipase levels were considered adequate laboratory markers for pancreatitis monitoring. Despite not knowing whether our patient was hypertriglyceridemic, the propofol infusion was continued with the focus on achieving optimum ventilatory goals given the failure of the other pharmacologic choices.

A question may arise as to whether the patient experienced a benzodiazepine or opiate withdrawal reaction based on continuous-infusion exposure. The infusions of morphine sulfate and lorazepam were discontinued approximately 2 and 3 weeks, respectively, before the withdrawal symptoms occurred. Based on the pharmacokinetics of both agents and the normal renal and hepatic function in our patient, the benzodiazepine or opiate withdrawal symptoms would have occurred earlier. Finally, the withdrawal symptoms abated approximately 30 minutes after restarting the propofol infusion and increasing the dosage on two separate occasions. This strengthens our hypothesis that the symptom complex observed in our patient represents a propofol withdrawal reaction.

On BTC day 108, the clinical pharmacist noted that the patient was being treated with a prolonged (79 days) infusion of propofol and initiated a plan for tapering and discontinuing the infusion. This intervention resulted in the first propofol withdrawal, the sudden onset of withdrawal symptoms, the apparent repeated requirement for propofol, and our search of the medical literature concerning dependence and withdrawal. The manufacturer of the propofol product, Diprivan (AstraZeneca Pharmaceuticals, Wilmington, DE), was also contacted to seek advice on how to withdraw the drug. Based on the collective experience from unpublished reports of propofol withdrawal in adult and pediatric patients, the propofol infusion was reduced by increments of 0.25 µg/kg/minute from an initial rate of 5 µg/kg/minute to a final rate of 4.5 µg/kg/minute. This incremental dosage reduction of 0.25 µg/kg/minute or 0.1 ml/hour was accomplished using a pediatric infusion pump. Weaning the patient at this very slow rate was proceeding with success and absence of withdrawal symptoms until his final severe sepsis episode with P. aeruginosa pneumonia and bacteremia occurred. This episode progressed to septic shock, multiple organ failure, and death.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: