Treatment of opioid-induced delirium with acetylcholinesterase inhibitors: a case report

1. Introduction 

Delirium occurring in the setting of advanced cancer is typically multifactorial in origin, with one common precipitating factor being the escalation of opioid dose escalation in patients with intractable pain.1., 2., 3. For the management of opioid-related delirium, a number of strategies have been recommended.4., 5., 6., 7. These include the recognition, and, when possible, reversal, of metabolic derangements, opioid rotation, the elimination of unnecessary adjuvant analgesics, and hydration. Unfortunately, these approaches are inadequate in some patients to reverse the disturbances of consciousness and cognitive function which characterize delirium.

Sedation is frequently a presenting sign of delirium in the opioid-treated patient. Recent reports have highlighted the value of donepezil, an oral centrally-acting acetylcholinesterase inhibitor, for the treatment of opioid-induced sedation.8., 9. Physostigmine, an injectable cholinesterase inhibitor, has previously been reported to reverse sedation in the postoperative opioidtreated patient and to reverse the delirium resulting from anticholinergic medications.10., 11. We wish to report the case of a patient having severe pain due to advanced ovarian cancer, whose opioid-associated delirium and myoclonus were promptly and significantly improved by an injection of physostigmine, and who maintained this benefit when placed on donepezil.

2. Case report 

The patient was a 55-year-old white woman with recurrent ovarian cancer and peritoneal metastases complicated by a chronic partial small bowel obstruction. Her medications were transdermal fentanyl 125 μg/hr applied every three days, oral morphine 30 mg as needed, alprazolam 0.5 mg as needed, gabapentin 300 mg three times daily, fluoxetine 80 mg daily, and trazodone 100 mg nightly. Her admission to the hospital had been precipitated by increasing abdominal pain, fever associated with an abdominal wall abscess, and dehydration. Routine chemistries were normal except for albumin 1.9, potassium 2.2, and magnesium 1.4. On admission her opioid consumption in oral morphine equivalents (OME) was approximately 300 mg/day. Her medications other than fluoxetine were continued, and cephalexin and levofloxacin were started empirically as antibiotic treatment. Consultation for severe pain and nausea one day following admission revealed no difficulties in arousal or mentation, and no evidence of myoclonus or asterixis. Her opioid management was changed to include morphine 4 mg/hr with intermittent bolus doses. A laxative regimen was begun for problems of constipation.

Five days later, she had developed progressively worsening problems of myoclonus, confusion, and sedation. These were coincident with an increase in her transdermal fentanyl to 175 μg/hr and continuing use of morphine for breakthrough pain (total OME 500 mg/day). Other medications were unchanged, although clonazepam 0.25 to 0.5 mg nightly had been started and alprazolam was no longer being requested. There was no change in her laboratory parameters. On examination, mentation was considerably slowed. She was disoriented except to situation, disorganized in her thinking, and delusional. Her ability to attend and shift attention was poor. Questions often had to be repeated three or four times before she would answer, and when she did respond, her answers more often referred to questions previously asked. The same pattern of distraction applied to her ability to follow simple commands and her spontaneous speech tended to perseveration. Her Memorial Delirium Assessment Scale (MDAS) score12 was 24. Myoclonus was multifocal and at times accentuated by movement. On testing for asterixis she could barely dorsiflex her wrists without having them flop downward. The impression was of a severe, hypoactive delirium.

Treatment considerations included modifications to her opioid and adjuvant medication regimen. Because the patient had taken multiple different opioids (e.g., hydromorphone, fentanyl, morphine) for treatment of her pain over the previous year, additional opioid rotation was not attempted. None of her adjuvant medications were new to her treatment nor had their dose been increased before the onset of her delirium. Based upon our previous experience with acetylcholinesterase inhibitors in treating sedation, we decided to try intravenous physostigmine to see if this would improve her mentation. After glycopyrrolate 0.1 mg was given to block peripheral cholinergic receptors, 1 mg of physostigmine (18 μg/kg) was intravenously infused over the course of 5 minutes. No change in her vital signs was observed. Re-examination five minutes later revealed the complete resolution of her asterixis and myoclonus. She became responsive to questions, answering all promptly and appropriately, and she followed simple commands without hesitation or error. Her husband described the improvement in her condition, as “dramatic” and “miraculous.” Re-evaluation 75 minutes later showed mentation to be less dramatically improved and asterixis of mild severity had returned. Fluoxetine, which had been omitted from the admission orders, was resumed and donepezil started at 5 mg/day. The following day she was more alert and appropriate, responded more quickly to questions, and was more consistent in following simple commands. Asterixis, however, was again severe. Her dose of donepezil was increased to 5 mg twice daily. The following day she had improved further; her MDAS score was 11. She was continued on transdermal fentanyl and 5 days later, at the time of discharge to hospice, her daily OME continued in the range of 500–600 mg/day. Nevertheless, her mentation remained improved and she did not have myoclonus or asterixis. Discharge medications included donepezil 5 mg twice daily and glycopyrrolate orally 1 mg three times daily. The patient's husband was interviewed four days later and reported no return of myoclonus or confusion, and the patient continued on donepezil.

3. Discussion 

Our patient had the characteristic features of delirium, namely, fluctuating levels of consciousness, inattention, and cognitive dysfunction. As occurs in many patients with delirium, multiple factors were identified that may have contributed to her condition. These included infection, polypharmacy, multiple metabolic derangements, and dehydration. The majority of these were under control at the time that her delirium first appeared. The inadvertent withdrawal of fluoxetine at time of admission, while another possible etiologic factor, was an unlikely precipitant due to the long half-lives of both the parent compound and its metabolite desmethylfluoxetine.13 The factor that best appeared to explain her delirium was the increase in her opioid dose.

When delirium develops in a patient on a high opioid dose, the most appropriate management strategy is usually a trial of rotation to an alternative opioid.6., 7., 14. Certain factors weighed against initiating an opioid rotation as our first management strategy in this patient. She had already been exposed to multiple opioids over her treatment course, thus somewhat limiting the options for opioid rotation. In view of her partial small bowel obstruction, we wanted to maintain her, if at all possible, on transdermal fentanyl. Had bowel obstruction not been a factor, therapeutic rotation to oxycodone would have been a reasonable management choice. We have generally avoided the practice of rotation to methadone in delirious patients, given the unpredictable pharmacokinetics of this agent, and its complex neuropharmacology. Had this patient not responded to the treatment given, opioid rotation to one of these two agents would have been tried.

Our patient's hypoactive delirium promptly and significantly improved following the administration of physostigmine. Within minutes her severe myoclonus and asterixis had markedly resolved, she became attentive and able for the most part to accurately respond to questions. Seventy-five minutes following the injection, this improvement had somewhat dissipated, the period of benefit matching the duration of physostigmine's expected central activity.15 Based upon the patient's improvement with physostigmine, treatment was initiated with donepezil, an oral acetylcholinesterase inhibitor approved for Alzheimer's disease. Her improvement on donepezil was dose dependent, and at the 10 mg daily dose she was able to maintain a level of improvement comparable to that which she had enjoyed short-term with physostigmine.

The value of the acetylcholinesterase inhibitors physostigmine and donepezil in reducing this patient's delirium speaks to the central role played by acetylcholine in modulating cortical arousal and information processing.16., 17. Cholinergic projections from the basal forebrain to the thalamus and cortex are considered essential in maintaining a state of conscious awareness and attention.17., 18., 19. Acetylcholine is also an important neurotransmitter in the maintenance of attention, and this process is impaired early in the evolution of delirium. Anticholinergic agents, for example, have been demonstrated to cause impairments in vigilance in excess of problems produced in memory function.20., 21. Hallucinations and other neuropsychiatric disturbances, so common in patients with delirium, may occur with disruption of the basal forebrain cholinergic pathways or in states of acetylcholine deficiency.22 All four acetylcholinesterase inhibitors approved by the U.S. Food and Drug Administration for Alzheimer’s disease, a condition marked by prominent intracerebral acetylcholine deficiency, have shown activity in decreasing neuropsychiatric and behavioral disorders in affected patient populations.23., 24., 25., 26.

Disturbances of the normal sleep-wake pattern account for some of the fluctuations in consciousness seen in delirious patients.27 Such disturbances in arousal and sleep may also in part be explained by deficiencies in cholinergic control. Cholinergic projections which extend from the pedunculopontine tegmental nucleus and laterodorsal tegmental nucleus to the thalamus and medial pontine reticular formation are important in generating and maintaining rapid eye movement (REM) sleep.28., 29. The injection of cholinergic agents into these pontine nuclei can promote REM sleep, whereas the opposite occurs following the injection of anticholinergic agents.30., 31. Along with converging inputs from the brainstem biogenic amine nuclei, pontine cholinergic projections contribute to the regulation of sleep-wake cycling, and to the overall state of cortical arousal.16., 32.

The established role of acetylcholine in mediating attention and cognition, coupled with our patient's prompt improvement after physostigmine, suggests that her delirium may have been related to the anticholinergic effects of her medications. The administration of anticholinergic medications is a well-recognized cause of delirium in both the young and old. Several reports do suggest that combinations of medications with low anticholinergic activity can have an additive effect and also cause anticholinergic toxicity.33., 34. A review of our patient's non-opioid medications does not disclose any that had major anticholinergic activity, and the number and dose of adjuvant medications having low anticholinergic activity would not seem to explain her delirium, particularly as the dose of these had been stable.

The only class of medications that were increased prior to the development of our patient's delirium was the opioids. There are a number of mechanisms through which opioids may impair cognitive processing, one of these being through the interference with central cholinergic transmission. Opioids demonstrate peripheral anticholinergic actions, and these are in part responsible for such opioid-related side effects as dry mouth, constipation, and urinary hesitancy. Both electrophysiological and behavioral studies indicate that morphine and other opioids can also inhibit central cholinergic activity in multiple cortical and subcortical regions of the brain.35., 36., 37., 38., 39., 40. Through inhibition of acetylcholine release in the medial pontine reticular formation, morphine has been shown to decrease REM sleep.29., 41. This can have disruptive effects of a more general nature on the normal sleep architecture,42., 43. which can, in turn, affect states of arousal during wakefulness.

If disturbances in central cholinergic transmission are in part responsible for opioidinduced delirium, then treatments which enhance cholinergic activity might prove palliative. The class of agents principally used to increase intracerebral cholinergic function is the centrally-active cholinesterase inhibitors. One of the longest used medications in this class is physostigmine, which has been shown to diminish attentional deficits caused by cholinergic dysfunction,44., 45. to heighten alertness when injected during the wake state,46 and to reverse anticholinergic delirium47., 48. (hence its original trade name, Antilirium). In opioid-associated delirium, symptoms often evolve along a continuum extending from sedation to a more global disturbance of consciousness and cognition. Physostigmine has been shown to reverse opioid-related sedation without causing other symptoms of opioid withdrawal, such as increased pain.10., 49., 50. Donepezil has likewise been reported to reverse opioid-induced sedation, and may be effective in treating associated symptoms of myoclonus and delirium.8., 9.

On the basis of these observations, it is reasonable to consider further evaluation of the acetylcholinesterase inhibitors in the treatment of patients whose delirium may be attributable to opioids. Given the recent literature suggesting that cholinergic augmentation can enhance analgesia,10., 50., 51., 52., 53., 54. acetylcholinesterase inhibitors may be neurochemically well positioned for this purpose.

In summary, opioid-induced encephalopathy, at first characterized by sedation, and when fully evolved, by delirium, may be due to the multiple inhibitory effects of opioids on cerebral cholinergic activity. Severe opioid-induced sedation is often an early sign of delirium, and has been successfully treated with acetylcholinesterase inhibitors. The present report suggests that the acetylcholinesterase inhibitors should be further evaluated as a treatment for delirium arising from opioid therapy.