• 
• 

Three Part Question

In [patients presenting to the emergency department with anticholinergic toxidrome], does the administration of [physostigmine] compared to [any other treatment] reduces [length of stay or need for intubation]?

Clinical Scenario

A 16 year old male is brought to the emergency department with altered mental status, dry skin, dilated pupils, tachycardia and a core temperature of 40,2ºC. He is mildly agitated. His mother found him lying on the floor of the basement. Although she isn’t aware of him taking any drugs, a small bag of Datura stramonium seeds was found next to him. His EKG shows sinus tachycardia without QRS widening. You suspect an anticholinergic poisoning, and wonder if giving physostigmine would be beneficial for the patient.

Search Strategy

Search made on July 21th 2016

A) No BestBETs or critical appraisals were found on this topic.

B) The website www.clinicaltrials.gov was searched for ongoing trials on the subject. No trials were found.

C) The Cochrane Library was searched for reviews on the subject. No reviews were found.

D) MEDLINE using the PubMed interface (1900 – July 2016)

#1 – (intoxication) OR (toxicology) OR (overdose) // 207941 articles
#2 – emergency // 294009 articles
#3 – physostigmine // 7153 articles
#4 – #1 AND #2 AND #3 // 50 articles

After reviewing the articles, 2 relevant papers were found.

Furthermore, reference lists of the relevant papers were checked for potential studies. No new studies were found.

E) EMBASE using the Elsevier interface (1947 – June 2016)

#1 – ‘intoxication’/exp OR ‘intoxication’ // 373473 articles
#2 – ‘toxicology’/exp OR ‘toxicology’ // 393825 articles
#3 – ‘overdose’/exp OR ‘overdose’ // 360651articles
#4 – #1 OR #2 OR #3 // 726484 articles
#5 – ‘emergency’/exp OR ‘emergency’ // 460132 articles
#6 – ‘physostigmine’/exp OR ‘physostigmine’ // 12790 articles
#7 – #4 AND #5 AND #6 // 297 articles

After reviewing the articles, 3 relevant papers were found.

Furthermore, reference lists of the relevant papers were checked for potential studies. No new studies were found.

Search Outcome

313 papers were found (excluding duplicates). After reviewing the articles, 310 papers were excluded as they were found irrelevant to the question, of insufficient quality (consisting mostly of case reports, short case series or review articles) or because they were written neither in English nor in French. The remaining papers were included in this review (the 2 articles from search D were also found in search E).

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Burns, M.J., et al. 2000 United States	Patients referred to a university hospital toxicology consultation service with anticholinergic poisoning and treated with physostigmine, benzodiazepines or both (n=52)	Retrospective observational study of consecutive patients	Complication rate* (initial treatments)	Physostigmine vs. BZD 2/30 (6,7%) vs. 10/22 (45,5%) (p<0.002)	Relatively small study. No randomization. Unblinded. Retrospective study. Hard to extract rates from the presented data. High risk of selection bias. Premorbid conditions, poisoning severity and co-ingestions not available.
			Complication rate* (sole treatments)	Physostigmine vs. BZD 2/26 (8%) vs. 4/7 (57%) (p=0.01)
			Mean hospital length of stay (hours) (initial treatments)	Physostigmine vs. BZD 35,5 ± 24,7 vs. 47,9 ± 31,1 (p=0.15)
			Mean hospital length of stay (hours) (sole treatments)	Physostigmine vs. BZD 34,1 ± 26,3 vs. 56,4 ± 35,3 (p=0.14)
Salen, P., et al. 2003 United States	Consecutive patients presenting to an emergency department with confirmed Jimsonweed poisoning and anticholinergic symptoms (n=17)	Retrospective observational study of consecutive patients	Mean length of stay (hours)	Physostigmine + BZD vs. BZD 42 ± 22 vs. 28 ± 27 (p=0.45)	Retrospective study. No randomization. High risk of information and allocation bias. Unblinded. Small sample size. Only 18% (3/17) patients received physostigmine
Watkins, J.W., et al. 2015 United States	Patients presenting with an anticholinergic toxidrome who were seen by a toxicologist (n=815)	Retrospective analysis of an international toxicology registry	Intubation rate	Physostigmine vs. No physostigmine 6,4 vs 8,4% (OR 0.73, 95 % CI 0.38–1.45 p=0.38)	Retrospective analysis. Few demographic details on the studied patients. High risk of selection bias. Shows association, but not causality.
			Intubation rate	Physostigmine alone vs. No physostigmine 1,9 vs 8,4% (OR 0.21, 95 % CI 0.05–0.87 p=0.031)

Comment(s)

*Complications were defined as endotracheal intubation, aspiration pneumonia, rhabdomyolysis, delayed recovery, and ethanol withdrawal syndrome Anticholinergic poisoning is a hallmark presentation in clinical toxicology. It is associated with a wide variety of medications and botanicals. Physostigmine has been used as an antidote for patients presenting with this type of poisoning. During the 1970s, anticholinergic poisonings were very common in the emergency department (ED); so much so that many advocated for the addition of physostigmine to the “coma cocktail” for undifferentiated altered mental status[1]. In 1980, Pentel reported the death of two patients following the administration of physostigmine. Both of these patients were being treated for tricyclic antidepressant (TCA) poisoning[2]. Following this report, many recommended a more cautious use of the antidote. Nowadays, it is generally contraindicated in patients presenting an anticholinergic poisoning and signs of sodium-channel blockade. We wanted to evaluate the potential benefits of physostigmine as a treatment for anticholinergic poisoning with regards to length of stay and intubation rates. No prospective studies were identified on the subject. All papers included were non-randomized retrospective observational studies and were prone to bias. Except for the study by Watkins, they included few patients and were probably underpowered. Finally, the articles poorly presented the demographic data of the patients receiving physostigmine compared to the ones who didn’t, making it hard to compare for potentially confounding factors. This short review demonstrates that the current literature is very weak on the subject and prone to many biases.

Clinical Bottom Line

In patients presenting to the ED with anticholinergic poisoning, the use of physostigmine might reduce the intubation rate, but not the mean length of stay. Unfortunately, the current data is very weak and prone to bias. Further prospective randomized studies on the subject are needed to support its use in anticholinergic toxicity.

Level of Evidence

Level 3 - Small numbers of small studies or great heterogeneity or very different population.

References

1. Burns, M.J., et al. A comparison of physostigmine and benzodiazepines for the treatment of anticholinergic poisoning Annals of Emergency Medicine 2000. 34(4): p.374-381.
2. Salen, P., et al. Effect of physostigmine and gastric lavage in a datura stramonium-induced anticholinergic poisoning epidemic American Journal of Emergency Medicine 2003. 21(4): p.316-317.
3. Watkins, J.W., et al. The Use of Physostigmine by Toxicologists in Anticholinergic Toxicity Journal of Medical Toxicology 2015. 11(2): p.179-184.