Article Text

Download PDFPDF
National audit of antidote stocking in UK emergency departments
  1. James T Harnett1,
  2. Sheena Vithlani2,
  3. Shabnam Sobhdam2,
  4. James Kent3,
  5. Lindsay McClure4,
  6. Simon HL Thomas5,6,
  7. Paul I Dargan1,7
  1. 1 Department of Clinical Toxicology, Guy's and St Thomas' NHS Foundation Trust, London, UK
  2. 2 London Medicines Information Service, London North West Healthcare NHS Trust, London, UK
  3. 3 NHS Specialist Pharmacy Service, UK
  4. 4 NHS National Procurement, Edinburgh, UK
  5. 5 National Poisons Information Service (Newcastle Unit), Newcastle Hospitals NHS Trust, Newcastle, UK
  6. 6 Institute of Cellular Medicine, Newcastle University, Newcastle, UK
  7. 7 Faculty of Life Sciences and Medicine, King's College London, London, UK
  1. Correspondence to Dr James T Harnett, Clinical Toxicology, Guy's and St Thomas' NHS Foundation Trust, London SE1 7EH, UK; j.harnett{at}


Background Previous audits of antidote stocking in UK hospitals have demonstrated variable but improving compliance with joint Royal College of Emergency Medicine and National Poisons Information Service guidance on antidote availability in emergency departments. The guidance was updated in 2017.

Aim To provide a current picture of compliance with the 2017 antidote guidance and compare this to previous audits.

Methods Questionnaires were distributed to all hospitals in the UK with an emergency department via medicines information and regional pharmacy procurement networks. Data were collected on availability and stock levels of category A (immediately available) and category B (available within 1 hour) antidotes. Additionally, data were collected on holdings of category C (held supra-regionally) antidotes and arrangements for sourcing these if not stocked locally.

Results 233 hospitals were surveyed and 178 replies (76.4%) were received. There were 73 hospitals (41.7%) fully compliant with guidance for category A, 34 hospitals (19.1%) for category B and 18 hospitals (10.1%) for both categories A and B antidotes. Few hospitals stocked category C antidotes (1.1%–34.8%). Evidence of formalised regional holding arrangements for category C antidotes, as advised in the guidance, was noted in some areas but many regions remain without such agreements.

Conclusions Most hospitals remain not fully compliant with stocking recommendations for categories A and B antidotes, with limited recent improvement. Category C antidotes are stocked by few hospitals although awareness of where these can be sourced appears to be increasing. Emergency departments should review their antidote stocking arrangements to ensure compliance with guidance. Formal arrangements for stocking of the more rarely used category C antidotes at a regional level are also required, where not already in existence, in order to assure their availability in an equitable way across the country.

  • accident & emergency medicine
  • audit
  • clinical pharmacy
  • protocols & guidelines
  • audit, self-inspection

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Antidotes are an important tool in the management of poisoned patients, with timely administration being potentially lifesaving.1 2 Problems with availability of antidotes at the point of care have been reported worldwide with the potential for deleterious effects on patient outcomes.3–10 Guidance on antidote stocking is published in North America and the UK, and these emphasise the importance of timely availability.11–13 Poisoning is a common clinical issue—the UK National Poisons Information Service (NPIS) received 41 098 telephone enquiries and 628 659 accesses to the online poisons information database TOXBASE in the 2017–2018 year.14 The top five poisons for which enquiries were received were paracetamol, ibuprofen, codeine, concentrated surfactants (as found in fabric cleaning liquid tablets) and sertraline.

In the UK, the Royal College of Emergency Medicine (RCEM) and NPIS first issued national guidance on emergency department stocking of antidotes in 2006 and this has been revised on three occasions, most recently in 2017.12 These guidelines were developed to ensure appropriate and timely availability of antidotes in all hospitals for the optimum management of poisoned patients. Changes to the guidance in 2017 included the addition of Intralipid and idarucizumab, with the deletion of glyceryl trinitrate, isosorbide dinitrate, dimercaprol and penicillamine. Antidotes are divided into three categories based on the urgency with which they should be used: category ‘A’—should be immediately available within the emergency department, category ‘B’—should be available within 1 hour (ie, stored within the hospital) and category ‘C’—are suitable to be held within a geographical region so that they are available within a time frame of around 4–6 hours. In 2015, minimum holding quantities were added to recommendations for all antidotes in categories A and B—these are based on average doses required to initiate treatment.

Audits have been undertaken following the release of each update to assess changes in antidote holdings and guide future updates. These were carried out in 2007 (Nash, Dines & Dargan, unpublished), 201015 and 2014.16 The most recent audit in 2014 showed that, despite improvements, there remains substantial variability in the availability of antidotes, in particular those in category C. The aim of this study was to provide a current picture of compliance with the 2017 UK antidote guidelines.


The questionnaire used for the 2014 audit was updated to reflect current RCEM/NPIS antidote guidance, using tick box questions regarding the stocking and availability of all included antidotes. Additional details were sought regarding category C antidotes because results of the previous audit suggested that availability of antidotes in category C could not be assured in an equitable way across the country. For these, we enquired about the presence of formal regional holding arrangements and where these did not exist, how hospitals planned to source a category C antidote should it be needed.

The questionnaires were sent via email to chief pharmacists, procurement leads and/or medicines information teams at all hospitals with an emergency department in the UK between October and November 2018. In Scotland, support with disseminating the questionnaire was provided by NHS National Procurement. Follow-up reminders were sent via email monthly thereafter with all responses received by 7th February 2019 included in the analysis.

Responses were collated using an excel spreadsheet and separated by category for analysis. For each antidote, we recorded whether this was stocked in each hospital, its availability within the recommended time frame (for categories A and B) and whether recommended stock levels (for categories A and B) were held. If the antidote was not stocked, the estimated time taken to source this from another hospital was also recorded. Data were compared with the previous national audits carried out in 2010 and 2014.15 16 Data from the 2007 audit were not directly comparable due to differences in question phrasing and data collection.


There were 178 responses received from the 233 hospitals surveyed, giving a response rate of 76.4%, similar to that of the 2014 audit (78.6%) and broadly similar across the UK (England 79.2% (141/178), Scotland 67.9% (19/28), Wales 66.7% (10/15) and Northern Ireland 66.7% (8/12)).

Category A: should be immediately available in the emergency department

Table 1 shows results for the availability of category A antidotes. Table 2 shows whether correct stock levels were held. Both provide comparison to the 2010 and 2014 audits.

Table 1

Availability of category A antidotes

Table 2

Stocking of category A antidotes

While the majority of hospitals held every category A antidote, only 60 hospitals (33.7%) held every antidote for immediate availability within the emergency department at the recommended stock levels (tables 1 and 2), although this has improved slightly since the 2014 audit in which 24.3% held every antidote in category A. In all, 73 hospitals (41.0%) were fully compliant by holding the recommended range of cyanide antidotes and all other category A antidotes. Each category A antidote was held in some quantity by >90% of hospitals, with the exception of the cyanide antidotes and ViperaTAb (tables 1 and 2).

For cyanide antidotes, four are listed in the guidance with the recommendation to hold dicobalt edetate, hydroxocobalamin and sodium thiosulfate; sodium nitrite may be used if dicobalt edetate is not available. An appropriate range of cyanide antidotes compatible with this guidance was held by 118 (66.3%) hospitals. Since the 2014 audit, the proportion of hospitals stocking hydroxocobalamin for immediate availability has increased from 55.6% to 71.9%. Three (1.7%) hospitals reported not stocking any antidote for cyanide toxicity. The mean (range) time given to obtain a cyanide antidote not stocked from another pharmacy was 180 (45–2880) min.

Category B: should be available within 1 hour

In all, 34 hospitals (19.1%) stocked all category B antidotes for availability within 1 hour (usually within the pharmacy department), including at least one of fomepizole or ethanol. If phentolamine was excluded (because of current manufacturing and supply issues), this increased to 44 (24.7%). Each category B antidote was held in some quantity by >90% of hospitals, with the exception of cyproheptadine, phentolamine and pyridoxine (tables 3 and 4).

Table 3

Availability of category B antidotes

Table 4

Stocking of category B antidotes

Fomepizole is the preferred antidote for methanol or ethylene glycol poisoning, with ethanol recommended only if this is unavailable. Fomepizole was stocked in 152 hospitals (85.4%), an increase since 2010 and 2014 (16.8% and 73.4%, respectively). In all, 75 (42.1%) stocked both fomepizole and ethanol, 77 (43.2%) stocked fomepizole alone and 15 (8.4%) stocked ethanol alone. In total, 11 hospitals (6.2%) reported not stocking any antidote for toxic alcohol poisoning; although better than 2010 (15.3%), this has not changed since 2014 (6.5%).

Category C: suitable to be held supra-regionally

As expected, category C antidotes were held at fewer hospitals, ranging from two (1.1%) holding glucarpidase to 62 (34.8%) holding pralidoxime (table 5).

Table 5

Category C antidotes

In comparison with the 2014 audit, the proportion of hospitals providing information on the nearest source of a category C antidote has increased from 15.8%–54.6% to 58.0%–70.9%, dependent on the individual antidote, although the accuracy of the proposed source was not verified. The mean time reported to source a category C antidote was estimated at under 4 hours for all except glucarpidase, which is available on urgent request from the distributor at any time.

Reporting of sourcing arrangements for category C antidotes was stronger in areas with formalised and embedded regional holding arrangements, notably Yorkshire & Humber, Scotland and Northern Ireland, with the remainder of the UK not covered by such agreements. Some less formalised arrangements were noted around NPIS centres or hospitals with a toxicology department and 53 hospitals (29.8%) identified the NPIS or a hospital with a clinical toxicologist as a point of contact for discussing and obtaining category C antidotes.


This was the fourth national audit of antidote stocking in the UK over the last 15 years, but the first to include all parts of the UK and to collect data for all antidotes in the RCEM/NPIS guidance.

Although many antidotes had high compliance rates individually, overall, only 18 hospitals (10.1%) were fully compliant with the RCEM/NPIS guidance in having all category A and B antidotes available within the required time frame and at adequate stock. Compared to the 2014 audit, changes in stocking of antidotes are modest and compliance with the guidance remains inadequate. Timely access to antidotes in emergency situations is critical in the management of many poisoned patients. In view of clear national guidance, any incident arising from antidotes not being available when required would be difficult to defend.

Of note in category A, hydroxocobalamin holding has increased (from 55.6% to 71.9%) and in category B there is improved stocking of fomepizole (from 73.4% to 84.8%) with fewer hospitals stocking ethanol (from 72.2% to 50.6%). Otherwise, changes in the proportion of hospitals stocking individual antidotes are modest and in the single percentage figures; a minority of hospitals (41%) correctly stock all category A antidotes, and only 19.1% stock all those in category B.

Although stocking of cyanide antidotes has improved, deviation from the guidance was evident in many hospitals. Several advised of local purchasing decisions being made not to stock the recommended range or having the recommended stock split across several hospitals within a Trust. Although cyanide poisoning is rare, immediate antidote access is essential and therefore obtaining a cyanide antidote from another hospital would not be clinically appropriate in a patient with severe acute cyanide poisoning.17

Intralipid was added as a category A antidote in the most recent 2017 guidance update and is well stocked with 97.8% of hospitals now holding some quantity of this, with 83.7% having this immediately available in the ED. Similarly, idarucizumab, a reversal agent for the anticoagulant effect of dabigatran, was added to category B at the last update and is now stocked by 93.8% of hospitals, with 88.8% of hospitals having it available within 1 hour.

Awareness of the closest stock of category C antidotes has improved in comparison to 2014, with over two-thirds of hospitals reporting that they are able to suggest a nearest source. Outside of regions with formal holding arrangements however, it was unclear whether this information was accurate or up to date, so the apparent improvement should be treated with caution. Many hospitals were not aware of who to contact regarding category C antidotes with many stating this would be delegated to the on-call pharmacist but not providing information on whom they would call. Less than a third of hospitals stated they would call the NPIS or a clinical toxicology department for assistance with category C antidotes, as is recommended in the guidance.

Awareness around sourcing category C antidotes was notably stronger in hospitals that were part of a formal regional holding arrangement with the devolved nations and the north of England leading in this respect. This underlines the importance of formal regional holding arrangements to improve access to these rarely used, but essential antidotes, as recommended in the guidance.

Two category C antidotes have national holding arrangements, the details of which can be accessed via the NPIS and on TOXBASE. Botulinum antitoxin is supplied to 37 hospitals by Public Health England (PHE) for cases of treating cases of infective botulism. Additionally, due to its prohibitive cost and the rarity of the indication for use (severe methotrexate toxicity), glucarpidase is held centrally by the manufacturer’s supplier with urgent courier arrangements available to facilitate supply. A previous PHE arrangement to supply pralidoxime to 95 hospitals is no longer operational; however, many hospitals still maintain stocks of this. Finally, separate national arrangements exist for the emergency distribution of antidotes in case of a major incident arising from deliberate release of a chemical or biological agent,18 and for the supply of anti-venoms for exotic snakes, spiders and other species.

Data from other countries including Australia, Canada, Ireland, New Zealand and the USA have shown similar problems with insufficient stocking of antidotes.4–11 Reasons for inadequate stocking were not investigated as part of this audit and no research has yet been directed at this question in the UK. Research in Canada has suggested several factors, including difficulty in acquisition, cost, infrequent usage and the lack of sharing arrangements with other hospitals.19–21 These factors are probably also relevant in the UK.


There are several factors which affect comparison of this study with previous audits. All countries in the UK were included this time, with Scotland not surveyed in 2014 and Northern Ireland not surveyed in 2010 due to other local audits being undertaken concurrently. Additionally, the 2010 audit did not include all antidotes recommended in the guidance, but chose a sample of 29 by consensus. Finally, there have been several additions and deletions to the guideline since the last audit.

Measurement of ViperaTAb holdings at correct stock levels was complicated by a reformulation of this antidote since the guidance was last updated, from 10 mL to 4 mL ampoules. Some hospitals may have reported incorrect stock despite holding adequate amounts of the new formulation, resulting in under reporting of adherence to the guidance for this antidote. There were supply problems reported for phytomenadione (vitamin K) tablets during the audit period; however, as an alternative, the parenteral preparation can be administered orally.22Those hospitals reporting an adequate stock of the intravenous preparation were therefore recorded as meeting the audit standard. Macrogol 3350 (polyethylene glycol) is used for whole bowel irrigation and is listed in the guidance with reference to a specific brand—Klean-Prep. This medication is also available from different manufacturers, for example as MoviPrep, so hospitals stocking this were recorded as meeting the audit standard. Finally, manufacturing and supply problems affecting phentolamine during the audit period will also have affected the results.

The overall response rate has slowly reduced between audits to 76.4% this year, perhaps reflecting the increasing workload on acute hospitals; it is possible that hospitals that did not respond have worse compliance with the guidance than responders and so the results may over-estimate antidote stocking nationally. There is also a risk that respondents to the survey were not counting actual stock on the shelf but instead reporting holdings based on electronic ordering systems. Respondents were not asked to comment on whether antidote stock they were holding was in date or not. Asking for this information would provide a more accurate picture of viable antidote stock holdings but might also affect response rates. Future audits should consider whether or not to request these data from respondents.


Compliance with national antidote stocking guidance remains inadequate. Further work is required to ensure emergency departments have timely access to the full range of antidotes required to provide effective treatment for severely poisoned patients.

Work is also needed to develop formalised, collaborative regional antidote stocking arrangements for the category C antidotes and supporting information on how to access these, where not already in existence.

What this paper adds

What is already known on this subject

  • The Royal College of Emergency Medicine (RCEM) and National Poisons Information Service (NPIS) issue guidance on the stocking of antidotes for the management of poisoned patients and these were last updated in 2017.

  • Previous audits of antidote availability in UK hospitals have demonstrated variable compliance with the RCEM/NPIS guidance.

  • Research in other countries has shown similar problems with adequate stocking of essential antidotes.

What this study adds

  • Compared to the 2014 audit, changes in stocking of antidotes are modest and compliance with the guidance remains inadequate.

  • There has been good uptake of antidotes added to the 2017 update (Intralipid and idarucizumab), with both now stocked in over 90% of hospitals.

  • Awareness of how to source regionally held category C antidotes has improved, though requires further work to establish formalised arrangements covering all of the UK.

  • Further work is also required to ensure emergency departments have timely access to the full range of antidotes required to provide effective treatment for severely poisoned patients.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements



  • EAHP Statement 2: Selection, Procurement and Distribution. EAHP Statement 4: Clinical Pharmacy Services.

  • Contributors JH made primary contributions to data collection and analysis, interpretation of results and writing of the manuscript. SV, SS, JK and LM assisted with data collection. All authors contributed to interpretation of results and revision of the manuscript, and all approved the final manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.