Physostigmine for Management of Anticholinergic Toxidrome

Authors: Sahaphume Srisuma, MD and James Dazhe Cao, MD (Rocky Mountain Poison and Drug Center) // Editor: Alex Koyfman, MD (@EMHighAK, EM Attending Physician, UTSW / Parkland Memorial Hospital) & Justin Bright, MD (@JBright2021)


Physostigmine, a carbamate extracted from Physostigma venenosum (Calabar bean), is a reversible acetylcholinesterase inhibitor that increases synaptic acetylcholine at both nicotinic and muscarinic receptors. Physostigmine’s primary therapeutic role aims to ameliorate delirium as a result of the anticholinergic (more accurately, antimuscarinic) toxidrome resultant from the blockade of muscarinic receptors by agents such as atropine, antihistamines, tricyclic antidepressant (TCA), amongst other xenobiotics. It can also be used diagnostically for undifferentiated altered mental status where anticholinergic delirium is suspected. Although physostigmine at one point historically was given liberally as part of the “coma cocktail”, more recent concerns regarding adverse events have made clinicians wary of its use. Knowledge of the evidence of physostigmine’s efficacy and adverse effects may ease hesitancy over the use of physostigmine.



Successful treatment of antimuscarinic toxicity by the use of physostigmine is limited to case series and reports. No randomized controlled trials have been conducted in humans demonstrating the efficacy of physostigmine. There are two retrospective chart reviews that evaluated the efficacy and adverse effects of physostigmine treatment.

Dr. Burns et al. in 1999 conducted a retrospective chart review comparing responses to physostigmine versus benzodiazepine in 52 antimuscarinic poisoning patients with delirium or hallucination. The authors found that physostigmine controlled agitation in 96% of patients as compared to only 24% of patients treated with benzodiazepines. Physostigmine also reversed delirium in 87% of patients, whereas none with benzodiazepines. Mean response time to physostigmine was 10.9 ± 5.3 minutes. Of the cases with response to physostigmine, 78% of patients had relapse of symptoms with a mean time of 100 ± 42 minutes. All patients receiving physostigmine received multiple doses. Time to recovery without further relapse was shorter in cases using physostigmine, but there was no significant difference in overall hospital length of stay.[1]

Schneir et al. in 2003 reviewed retrospectively 39 cases where physostigmine was administered diagnostically in cases of delirium.

 Adverse Effects and Toxicities

The principal adverse effects of physostigmine are related to cholinergic excess including bradycardia, bronchospasm, bronchorrhea, seizure, and motor weakness. Less severe symptoms are nausea, vomiting, diarrhea, miosis, tremor, and fasciculation.

The most concerning and also controversial adverse effects of physostigmine are bradydysrhythmias and asystole in case with use for TCA poisoning. Widespread concern for the use of physostigmine followed the landmark series of two cases by Drs. Pentel and Peterson in 1980. One case was a severe amitriptyline poisoning presented with alteration of consciousness with response only to deep pain. After intubation, patient developed status epilepticus. Initial EKG showed very wide QRS complex (approximately 240 msec) and first degree AV block. After administration of physostigmine 2 mg IV over three minutes, patient developed nodal bradycardia and further asystole. The second case was of imipramine and propranolol poisoning that had two episodes of seizure and hypotension. Physostigmine 2 mg IV was given over five minutes after each episode of seizure. After second dose of physostigmine, patient developed bradycardia and progressed to asystole. The authors concluded that the “use of physostigmine in patients who have ingested an overdose of TCAs carries the risk of life-threatening bradyarrhythmias.”[11] Although the authors correctly noted that bradycardia is uncommon in TCA overdose, bradydysrhythmias and asystole may be terminal events in the setting of severe TCA toxicity.[10] NaHCO3 was not given to treat sodium channel blockade effect of severe TCA toxicity prior to cardiac arrest. Additionally, underlying conduction delay demonstrated by the first degree AV block in case one and co-ingestion of propranolol in case two may have contributed to the adverse outcomes. Even so, the case series yielded significant concern for the use of physostigmine in the setting of TCA overdose, QRS prolongation, and/or PR prolongation.

In Burns’ 1999 publication, there was no difference in side effects defined as complications arising within 30 minutes of medication administration and between cases treated with physostigmine and with benzodiazepine. However in this study, PR prolongation (>200 msec) or QRS widening (>100 msec and not related to bundle branch block) were considered as contraindication to physostigmine. Side effects were reported in five of 45 cases with physostigmine – one of each of the following: diaphoresis, emesis, diarrhea, asymptomatic bradycardia @ 51 beats/minute, and increased respiratory secretions. In the 25 cases with benzodiazepine, there were side effects reported in four cases including two with excessive sedations, one fecal incontinence, and one paradoxical agitation. There were five cases with TCA poisoning in this series, all had ingested amitriptyline at least 12 hours before physostigmine administration. None had coma, seizures, hypotension, cardiac conduction disturbances, or significant dysrhythmias although significant TCA toxicity would have likely been excluded from the study.[1]

In Schneir’s series, one of 39 cases had brief convulsions without adverse sequelae after treatment with physostigmine. The patient ingested doxylamine and presented after a witnessed 1-2 minutes of seizure. He was later suspected to have antimuscarinic delirium for which he received physostigmine 0.5 mg IV with full reversal of delirium. Twelve minutes after administration of physostigmine, the patient had a 30 second generalized convulsion without further convulsions or sequelae during the remainder of the hospitalization. EKG was available in 37 cases. Only one case had QRS > 120 msec (138 msec) which was known to be a preexisting right bundle branch block. Of the three TCA cases in the series, none of the patients had dysrhythmias. There were no cases of cholinergic excess following physostigmine administration.

Three other case series have described incidence of seizures temporally after physostigmine administration. Newton et al. in 1975 reported convulsions in two of 21 patients treated with physostigmine for TCA overdoses.[12] Walker et al. in 1976 described convulsions in three out of 26 overdose patients treated with physostigmine. Of the 26 patients, 17 were reported to have ingested TCAs.[13] Knudsen et al. in 1984 described a series of 41 patients after ingestion of maprotiline where six of seven patients treated with physostigmine developed convulsions.[10]

Discussion and Recommendation

            From the understanding of physostigmine’s mechanism of action and the above data, we would like to summarize key points and recommendations regarding the use of physostigmine. First, physostigmine is beneficial in reversing central antimuscarinic symptoms including agitation, delirium, and/or hallucinations. When properly administered, physostigmine may be more efficacious than benzodiazepines for management and lowers risk of excessive sedation from high doses of benzodiazepine. Physostigmine can also be applied diagnostically for undifferentiated delirium; preventing unnecessary investigation, especially in children.

Second, physostigmine will NOT treat symptoms from pathophysiology other than antimuscarinic effects. Without any antimuscarinic symptoms, do NOT use physostigmine to treat unspecified coma or dysrhythmias. Physostigmine may help with antimuscarinic effect of tricyclic antidepressant but is unlikely to reverse serotonergic effect or alpha adrenergic blockade of TCAs. Physostigmine has no direct reversing properties for sodium channel blockade in TCAs. Theoretically by slowing the heart rate, physostigmine may improve cardiac sodium channel blockade. However, data for this hypothesis is virtually nonexistent in humans and conflicting in animal studies. Therefore in cases of TCA overdose, it is very important to focus on which clinical symptoms are significant and need to be treated.

Third, when using physostigmine, understand the risks and adverse effects. The concerning adverse effects of cholinergic excess include seizure, bradycardia, bronchospasm, and bronchorrhea. We recommend AGAINST using physostigmine for treatment of seizures caused by an anti-muscarinic agent. Benzodiazepine may be the most appropriate first choice. We also recommend AGAINST using physostigmine in cases with bradycardia or AV block. The use of physostigmine for cases with QRS widening remains controversial. We recommend focusing on therapies targeted at reversing cardiac sodium channel blockade by either administration of NaHCO3 or hypertonic saline.

Dosing Recommendations

We recommend 1-2 mg (0.05 mg/kg, maximum initial dose of 0.5 mg in children) IV slow infusion over at least 5 minutes to reduce the risk of seizures. Dose may be repeated for incomplete response after 5 to 10 minutes up to a maximum of 2 mg in children and 4 mg in adults. Always have atropine and a benzodiazepine at bedside in case of significant cholinergic excess symptoms or seizures, respectively. Response to physostigmine can occur rapidly (within minutes), but the duration of effect of physostigmine tends to be shorter than that of antimuscarinic agents. Observe for recurrence of antimuscarinic symptoms, and assess if the patient will need repeated dosing of physostigmine.


Physostigmine is beneficial for central and peripheral antimuscarinic symptoms. With appropriate use, it has a low risk of side effects and complications.


References / Further Reading

[1] Burns MJ, Linden CH, Graudins A, Brown RM, Fletcher KE. A comparison of physostigmine and benzodiazepines for the treatment of anticholinergic poisoning. Annals of emergency medicine. 2000;35:374-81.

[2] Schneir AB, Offerman SR, Ly BT, Davis JM, Baldwin RT, Williams SR, et al. Complications of diagnostic physostigmine administration to emergency department patients. Annals of emergency medicine. 2003;42:14-9.

[3] Glatstein MM, Alabdulrazzaq F, Garcia-Bournissen F, Scolnik D. Use of physostigmine for hallucinogenic plant poisoning in a teenager: case report and review of the literature. American journal of therapeutics. 2012;19:384-8.

[4] Johnson PB. Physostigmine in tricyclic antidepressant overdose. Jacep. 1976;5:443-5.

[5] Rumack BH. Anticholinergic poisoning: treatment with physostigmine. Pediatrics. 1973;52:449-51.

[6] Phillips MA, Acquisto NM, Gorodetsky RM, Wiegand TJ. Use of a physostigmine continuous infusion for the treatment of severe and recurrent antimuscarinic toxicity in a mixed drug overdose. Journal of medical toxicology: official journal of the American College of Medical Toxicology. 2014;10:205-9.

[7] Padilla RB, Pollack ML. The use of physostigmine in diphenhydramine overdose. The American journal of emergency medicine. 2002;20:569-70.

[8] Cole JB, Stellpflug SJ, Ellsworth H, Harris CR. Reversal of quetiapine-induced altered mental status with physostigmine: a case series. The American journal of emergency medicine. 2012;30:950-3.

[9] Weizberg M, Su M, Mazzola JL, Bird SB, Brush DE, Boyer EW. Altered Mental Status from Olanzapine Overdose Treated with Physostigmine. Clinical toxicology. 2006;44:319-25.

[10] Knudsen K, Heath A. Effects of self poisoning with maprotiline. Br Med J (Clin Res Ed). 1984;288:601-3.

[11] Pentel P, Peterson CD. Asystole complicating physostigmine treatment of tricyclic antidepressant overdose. Annals of emergency medicine. 1980;9:588-90.

[12] Newton RW. Physostigmine salicylate in the treatment of tricyclic antidepressant overdosage. JAMA : the journal of the American Medical Association. 1975;231:941-3.

[13] Walker WE, Levy RC, Hanenson IB. Physostigmine–its use and abuse. Jacep. 1976;5:436-9.




Leave a Reply

Your email address will not be published. Required fields are marked *