TOXICOLOGY/CONCEPTS

Expert Consensus Guidelines for Stocking of Antidotes inHospitals That Provide Emergency Care

Richard C. Dart, MD, PhDStephen W. Borron, MD, MSE. Martin Caravati, MD, MPHDaniel J. Cobaugh, PharmDSteven C. Curry, MDJay L. Falk, MDLewis Goldfrank, MDSusan E. Gorman, PharmD, MSStephen Groft, PharmDKennon Heard, MDKen Miller, MD, PhDKent R. Olson, MDGerald O’Malley, DODonna Seger, MDSteven A. Seifert, MDMarco L. A. Sivilotti, MSc, MDTammi Schaeffer, DOAnthony J. Tomassoni, MD, MSRobert Wise, MDGregory M. Bogdan, PhDMohammed Alhelail, MDJennie Buchanan, MDJason Hoppe, DOEric Lavonas, MDSara Mlynarchek, MPHDong-Haur Phua, MDSean Rhyee, MD, MPHShawn Varney, MDAmy Zosel, MDFor the Antidote Summit Authorship

Group

From the Rocky Mountain Poison & Drug Center - Denver Health, Denver, CO (Dart, Heard, Schaeffer,Bogdan, Alhelail, Buchanan, Hoppe, Lavonas, Mlynarchek, Phua, Rhyee, Varney, Zosel); Department ofSurgery, University of Texas Health Sciences Center, San Antonio, TX (Borron); Division of EmergencyMedicine, Utah Poison Control Center, University of Utah Health Sciences Center, Salt Lake City, UT(Caravati); ASHP Research and Education Foundation, Bethesda, MD (Cobaugh); Department ofMedical Toxicology and Banner Poison Control Center, Banner Good Samaritan Medical Center,Phoenix, AZ (Curry); Department of Emergency Medicine, Orlando Regional Medical Center, University ofFlorida, Orlando, FL (Falk); New York City Poison Center; New York University School of Medicine, NewYork, NY (Goldfrank); Division of Strategic National Stockpile, Centers for Disease Control andPrevention, Atlanta, GA (Gorman); Office of Rare Diseases Research, Bethesda, MD (Groft); Division ofEmergency Medicine, School of Medicine (Dart, Heard, Lavonas, Schaeffer) and Department ofPharmaceutical Sciences, School of Pharmacy (Bogdan), University of Colorado Denver, Aurora, CO;Orange County Fire Authority and Orange County Health Care Agency Emergency Medical Services,Irvine, CA, and National Association of EMS Physicians, Lenexa, KS, (Miller); University of California,San Francisco, and San Francisco Division, California Poison Control System, San Francisco, CA(Olson); Division of Research, Department of Emergency Medicine, Albert Einstein Medical Center,Philadelphia, PA (O’Malley); Tennessee Poison Center, Division of Clinical Pharmacology, Department ofMedicine, Vanderbilt University Medical Center, Nashville, TN (Seger); University of New Mexico Schoolof Medicine & Medical Director, New Mexico Poison and Drug Information Center, Albuquerque, NM(Seifert); Department of Emergency Medicine and Department of Pharmacology & Toxicology, Queen’sUniversity, Ontario, Canada (Sivilotti); Yale University School of Medicine, Department of Surgery,Section of Emergency Medicine, and Yale-New Haven Center for Emergency Preparedness and DisasterResponse, New Haven, CT (Tomassoni); Division of Standards and Survey Methods, The JointCommission, Oakbrook Terrace, IL (Wise).

Study objective: We developed recommendations for antidote stocking at hospitals that provide emergency care.

Methods: An expert panel representing diverse perspectives (clinical pharmacology, clinical toxicology, critical caremedicine, clinical pharmacy, emergency medicine, internal medicine, pediatrics, poison centers, pulmonary medicine, andhospital accreditation) was formed to create recommendations for antidote stocking. Using a standardized summary of themedical literature, the primary reviewer for each antidote proposed guidelines for antidote stocking to the full panel. Thepanel used a formal iterative process to reach their recommendation for the quantity of an antidote that should be stockedand the acceptable period for delivery of each antidote.

Results: The panel recommended consideration of 24 antidotes for stocking. The panel recommended that 12 of theantidotes be available for immediate administration on patient arrival. In most hospitals, this period requires that theantidote be stocked in the emergency department. Another 9 antidotes were recommended for availability within 1 hour ofthe decision to administer, allowing the antidote to be stocked in the hospital pharmacy if the hospital has a mechanism forprompt delivery of antidotes. The panel identified additional antidotes that should be stocked by the hospital but are notusually needed within the first hour of treatment. The panel recommended that each hospital perform a formal antidotehazard vulnerability assessment to determine the need for antidote stocking in that hospital.

Conclusion: The antidote expert recommendations provide a tool to be used in creating practices for appropriate andadequate antidote stocking in hospitals that provide emergency care. [Ann Emerg Med. 2009;54:386-394.]

0196-0644/$-see front matterCopyright © 2009 by the American College of Emergency Physicians.doi:10.1016/j.annemergmed.2009.01.023

386 Annals of Emergency Medicine Volume , . : September

Dart et al Antidote Stocking Guidelines

INTRODUCTIONAntidotes are a critical component in the care of poisoned

patients. Antidotes such as digoxin immune Fab can belifesaving; however, an antidote must be available at theappropriate time to be effective. For some poisons, the antidotemust be available immediately. The administration of cyanideantidote can resuscitate a patient only if the antidote isadministered before irreversible injury develops. For otherantidotes, there is time to procure the drug from the pharmacyor from another hospital. The use of a specific antidote wasreported by US poison centers approximately 80,000 times in2006.1 Unfortunately, important antidotes often are notstocked at all or are stocked in an insufficient amount.Insufficient stocking of a diverse group of antidotes has beendocumented repeatedly in many countries, including the UnitedStates and Canada.2-9

Although national recommendations of an expert panel werepublished in 2000, reports of inadequate stocking persist.10,11

The causes of this serious problem are unknown but are likelyrelated in part to limited awareness, infrequent use,interruptions in supply, and allocation of limited hospitalpharmacy resources. Previous studies have found that largerhospitals are more likely than smaller or rural hospitals to stockantidotes adequately.4,7 Perceived cost of antidotes based onpurchase price, as well as pharmacist and physician unfamiliaritywith poisons and their antidotes, may also contribute.4,12

Changes in the types of antidotes available may also play a role.In recent years, new antidotes have become available and othershave been discontinued.

The Joint Commission (TJC) sets standards for accreditationof hospitals in the United States but does not explicitly addressantidote stocking. TJC standard MM.2.10 states simply thatmedications available for dispensing or administration beselected, listed, and procured according to criteria. StandardMM.2.30 states that emergency medications or supplies, if any,be consistently available, controlled, and secured.13 Individualstate governments also regulate hospitals, although theirattention in terms of antidotes has primarily been focused onmass casualty and terrorist events. However, California recentlysanctioned a hospital for violating a regulation requiring“. . .availability of prescribed medications 24 hours a day.”14 Inthat case, digoxin Fab was not available immediately for apatient with cardiac glycoside toxicity.

Although published documentation of insufficient antidotestocking is common, scholarly research and analysis of thephenomenon are almost nonexistent. Often, the efficacy of anantidote is well studied, but few studies address the number ofpatients or the period in which patients must be treated. Giventhe approval of new antidotes, the changes in availability ofantidotes, a changing regulatory environment, and thecontinued lack of antidote stocking, the objective of thisevidence-based consensus process was to developrecommendations for the stocking of antidotes at hospitals that

provide emergency care. To produce useful and clinically

Volume , . : September

relevant guidelines despite an evidence base that is incomplete,we combined a structured analysis of the existing literature withan expert consensus panel.

MATERIALS AND METHODSOverview

Recommendations for antidote stocking were created in 2phases similar to the development of American College ofEmergency Physicians (ACEP) clinical policies. First,standardized evidence-based summaries of the medical literaturewere generated for each antidote. Each summary was thenindependently reviewed and revised by a primary reviewer fromthe expert panel. The reviewer presented the summary and therecommendation to the full panel, and an iterative process wasused to reach consensus. The panel was instructed to addressspecifically the needs of hospitals that provide emergency care inthe United States. Stocking of antidotes for mass casualty eventswas not addressed by the panel. Details about administration ofeach antidote were not addressed by the panel.

Phase 1Relevant medical literature was obtained by nonmedical staff

with extensive experience in searching and retrieving medicalliterature. Evidence-based summaries of the medical literaturefor each antidote were created by a group of emergencyphysicians and clinical toxicologists not involved in theconsensus process. For each antidote, a standardized summaryof 5 to 20 pages was created for subsequent assessment by theprimary reviewer. Publications used to create the summary wereidentified with 3 methods: (1) searches for each antidote and itsindications, using the National Library of Medicine’s PubMeddatabase (http://www.pubmed.gov); limited to “human” and“English” for article types “clinical trials,” “reviews,” or “casereports”; (2) review of chapter bibliographies for each antidotein 2 textbooks of toxicology15,16; and (3) review ofbibliographies of selected articles from the previous 2 methodsfor additional citations. Each article was classified according toits methodology, using the clinical guideline model of theACEP (class I: good-quality randomized and blinded clinicaltrials and good-quality systematic reviews of good-qualityrandomized trials; class II: prospective, nonrandomized, ornonblinded clinical trials, cohort, or well-designed case-controlstudies, good-quality observational or volunteer studies; classIII: retrospective case series, case studies, relevant expertopinions, or animal studies) and then summarized with astandardized form.17

Phase 2Each literature summary developed in phase 1 was provided

to 1 expert panel member serving as the primary reviewer forthat antidote. The expert panel was a diverse group of 19professionals representing various perspectives (Table 1). Theprincipal investigator served as the nonvoting chairperson and

selected individuals for the panel according to evidence of

Annals of Emergency Medicine 387

Antidote Stocking Guidelines Dart et al

previous antidote research, professional experience with theacquisition and use of antidotes, or their potential role in theprovision of antidotes (eg, TJC). This approach was necessarybecause there is no formal body or compendium that evaluatescandidate antidotes or their appropriate stocking.

The primary reviewers assessed and revised the literaturesummary produced in phase 1 for each assigned antidote, usingtheir knowledge and experience. The primary reviewer couldadd articles to the summary. Each primary reviewer then formeda provisional recommendation about the antidote and presentedthe revised literature summary and the recommendation to theentire panel. The panel’s deliberations occurred on March 6 to7, 2008. The evidence-based analysis was formulated to provideinformation to the panel about the fundamental questionsinvolved in the selection of each antidote:1. Is the antidote effective?2. Do the medical benefits of its use outweigh its risks?

If the consensus was affirmative for the first 2 questions,the panel addressed 3 additional questions:

3. Is time an important factor in its use?a. Does the antidote need to be available immediately on

patient arrival in the emergency department (ED)?b. Does the antidote need to be available for administration

within 60 minutes of the decision to use?4. How many patients should a facility prepare for?5. What amount of the antidote is needed to treat 1 patient

weighing 100 kg?An iterative process was used to reach consensus on stocking

of each antidote. After presentation of an antidote by theprimary reviewer and discussion by the entire panel, a vote wastaken to determine consensus. An antidote was recommendedfor stocking if the panel consensus was affirmative for the first 2questions. The additional questions were addressed to assisthospitals in determining where an antidote should be stockedand in what quantity. For all questions, consensus was definedas agreement of at least 75% of eligible panel members,

Table 1. Profile of antidote panel members.

Discipline or Specialty No. of Participants

Clinical pharmacology 3Critical care medicine 3Clinical pharmacy 2Disaster preparedness/response 6Emergency medicine 11Emergency medical services 4Hospital pharmacy 1Internal medicine 2Clinical toxicology 15Pediatrics 2Poison center administration 9Public health 1Regulatory medicine or hospital accreditation 4

Categories were self-selected by panel participants. Total is greater than 19 be-cause of multiple designations of some individuals.

provided there was no strong disagreement vote. Each member

388 Annals of Emergency Medicine

could vote in one of 3 ways: agreement, disagreement, or strongdisagreement. If 1 or more panel members expressed strongdisagreement, the discussion was continued and another votetaken. If agreement could not be reached, the decision was listedas “consensus not reached.”

The panel’s estimate of the antidote amount needed perpatient was based on clinical considerations: dose, duration oftherapy (8 or 24 hours), use of extracorporeal elimination suchas hemodialysis, and other factors. The panel chose to considerone 100-kg patient as the basis for calculating the amount ofantidote to stock. This weight was chosen by using recent datafrom the National Health and Nutrition Examination Survey.18

This weight falls between the 75th and 85th percentile for menand approximates the 90th percentile for women.

The acquisition cost of each antidote was estimated bymultiplying the average wholesale price by the amount of drugrecommended by the panel. The average wholesale price is theestimated retail cost of a drug. In reality, most institutions havepurchasing agreements that allow them to purchase at a pricebelow average wholesale price. When a partial unit would beneeded to complete the dosing, the number of units (eg, vials)of the drug was rounded to the next full unit. For example, thecalculation of acetylcysteine based on weight usually results in apartial vial; therefore, cost of purchasing was estimated byrounding up to the next full vial.

Competing interests were managed proactively andtransparently. Each panel member completed a competinginterest form for each antidote, disclosing any financial interestor stock ownership or financial support (research grants,consulting agreements) from each antidote manufacturer ormarketer for the preceding 10 years. Any relationship (ie,funding for a clinical trial, a single consultation with thecompany, or any level of equity holding in the company) wasconsidered a competing interest. No participant reported equityholdings in a company. Six participants reported funding forclinical research and 6 reported previous consulting agreementswith an antidote manufacturer. The panel was informed of allcompeting interests for each antidote as it was considered. Eachpanel member with a competing interest was allowed toparticipate in discussion but was excluded from voting on theantidote involved.

RESULTSThere were 1,446 articles retrieved and reviewed; 583 articles

were used to develop the literature summaries and provisionalrecommendations. Class I evidence was occasionally available,typically to show the efficacy of the drug. Class II evidence wasmore commonly available but again focused on efficacy. ClassIII evidence was plentiful but extremely variable. A few class IIIstudies were rigorously performed medical record reviews, butmost were case reports or chart reviews without an appropriatedescription of methods. Virtually no articles explicitly addressedthe questions of the appropriate time for availability and

number of patients a facility should be prepared to treat.

Volume , . : September

Dart et al Antidote Stocking Guidelines

Overall, the panel recommended consideration of 24antidotes for stocking in hospitals that accept emergencypatients, counting only 1 antidote when alternatives wereavailable (eg, fomepizole or ethanol for treatment of toxicalcohols). The panel recommended that 12 of these antidotes beimmediately available for administration on patient arrival(Table 2). Antidotes for opioid, cardiac glycoside, cyanidepoisoning, or other conditions may be lifesaving if administeredbefore irreversible injury has occurred. In most hospitals, thisperiod requires that the antidote be stocked in the ED. Another9 antidotes were recommended for availability within 1 hour ofthe decision to use the antidote (Table 2), allowing the antidoteto be stocked in the pharmacy, providing the hospital has anefficient mechanism for prompt delivery of medications to theED. The panel recommended that 3 additional antidotes bestocked, but they are not necessarily needed within 1 hour ofordering. Consensus was not reached for Prussian blue. Finally,2 antidotes, both antitoxins for botulism, were notrecommended for stocking.

For some conditions, more than 1 antidote can effectivelytreat a poisoning or overdose. The panel identified 3 instancesin which more than 1 effective antidote was available: ethanol orfomepizole for treatment of toxic alcohol exposure, antiveninCrotalidae polyvalent immune Fab (ovine) (CroFab; FabAV) orantivenin (Crotalidae) polyvalent for treatment of crotalinesnakebite, and the conventional cyanide antidote kit (sodiumnitrite and sodium thiosulfate) or hydroxocobalamin for cyanidetoxicity. In these cases, the panel designated a preferred agent,although either agent was recognized as acceptable in meetingthe need for stocking. The preference was determined in thesame manner as the decision to recommend stocking of anantidote: by group debate reaching consensus without a vote ofstrong disagreement. Fomepizole was preferred over ethanol forseveral reasons: simplicity of use, lack of need for compoundingin pharmacy, reduction in medication errors, potential to avoidhemodialysis, and anticipated safety in children. The use ofethanol is further complicated by the lack of a commerciallyavailable 10% solution in the United States, requiringcompounding of a 10% solution from a 95% solution ofethanol.19 Hydroxocobalamin was preferred over theconventional cyanide antidote kit because of its widerindications, ease of use, and anticipated safety in widespreaduse. FabAV was preferred over antivenin (Crotalidae) polyvalentbecause of improved safety profile and the fact that productionof the Wyeth antivenom has been discontinued. Supplies of theWyeth antivenom were available but difficult to obtain at thetime of the panel’s assessment.

The panel recommended the amount of antidote needed totreat a 100-kg patient for either 8 hours or 24 hours (Table 3).In most cases, the amount of antidote recommended forstocking did not match the package label precisely because ofchanges in clinical practice since the label content was approved

by the Food and Drug Administration and because some

Volume , . : September

antidotes, such as octreotide and others, are not labeled for theiruse as an antidote.

The panel observed that many considerations can affect thedecision to stock an antidote, as well as the amount of anantidote that should be stocked. A rigid recommendation for allhospitals is difficult to justify and may lead to insufficientstocking. For example, a hospital in an area endemic forcrotaline snakes (rattlesnakes, copperhead snakes, watermoccasins) should stock antivenom, but the amountrecommended for 1 patient may be insufficient if 2 bites couldrequire treatment simultaneously. To address these situations,the panel recommended that hospitals perform a hazardvulnerability assessment for each antidote (Tables 3, 4).

LIMITATIONSLittle class 1 and 2 evidence was available for most antidotes;

therefore, many of the panel’s recommendations are based onexpert analysis and experience. The process attempted tocompensate for individual bias by using a large diversified panel,by presenting structured summaries of medical information, andby prohibiting voting by members with a competing interest.This approach helped to constrain undocumented orunsubstantiated opinion of panel members in 2 ways. First, thepublished medical evidence was reviewed and this supported theexpectation that the reviewer’s conclusions would be evidencebased within the limits of available information. Second, theother panel members (who reviewed the evidencesimultaneously) acted as a counterbalance againstunsubstantiated individual positions of the reviewer.

The panel was chosen by the nonvoting chairman accordingto documented clinical and research expertise, which may haveresulted in an unintended bias toward academia. This possibilitywas counterbalanced by the voting rules that allowed a singleobjection to reject a recommendation. Several exotic antidotesand antidotes not readily available in North America were notconsidered. The panel was also asked not to anticipate singularand rare events, such as terrorist acts or mass casualty incidents.It was not possible to assess the full cost-benefit relationshipbecause information about the value of benefit is not available.The intended audience of the recommendations was anindividual hospital providing emergency care, rather than largerregions, states, or national organizations, given the differentneeds and resources of such entities.

DISCUSSIONThe antidote expert panel recommended consideration of

24 antidotes for emergency stocking by facilities that provideemergency care. The recommendations are intended to beinterpreted in the context of the potential clinical uses createdby the catchment area served by a hospital; special needs formass casualty events are not addressed in theserecommendations.

Insufficient stocking of antidotes needed on an emergency basis

has been documented repeatedly in the United States and other

Annals of Emergency Medicine 389

Antidote Stocking Guidelines Dart et al

Table 2. Antidote recommendations for stocking at facilities that accept emergency patients.

Antidote Poisoning Indication(s)

Recommendation

Class ofEvidence

†Should BeStocked

AvailableWithin 60Minutes

ImmediatelyAvailable*

Acetylcysteine Acetaminophen Yes Yes No I (IV)II (oral)

Antivenin (Crotalidae) polyvalent,Wyeth, OR

North American crotaline snakeenvenomation

Yes Yes No III

Crotalidae Polyvalent ImmuneFab, ovine

‡North American crotaline snake

envenomationYes Yes No II

Antivenin (Latrodectus mactans) Black widow spider envenomation Yes No No IIIAntivenin (Micrurus fulvius) Eastern and Texas coral snake

envenomationYes Yes No III

Atropine sulfate Organophosphorus and N-methylcarbamate insecticides

Yes Yes Yes III

Botulism antitoxin, equine (A, B) Botulism No NA NA IIIBotulism immune globulin

(BabyBIG)Infant botulism No NA NA I

Calcium chloride§

Fluoride, calcium channelblocking agent

Yes Yes Yes III

Calcium gluconate§

Yes Yes Yes IIICalcium disodium EDTA Lead Yes No No IICalcium trisodium pentetate

(Calcium DTPA)Internal contamination with

plutonium, americium, orcurium

Yes No No III

Cyanide Antidote Kit OR Cyanide poisoning Yes Yes Yes IIIHydroxocobalamin hydrochloride

‡Cyanide poisoning Yes Yes Yes II

Deferoxamine mesylate Acute iron poisoning Yes Yes No IIDigoxin Immune Fab Cardiac glycosides/steroid

toxicityYes Yes Yes II

Dimercaprol Heavy metal toxicity (arsenic,mercury, lead)

Yes Yes No II

Ethanol§

OR Methanol, or ethylene glycolpoisoning

Yes Yes No III

Fomepizole‡

Methanol, or ethylene glycolpoisoning

Yes Yes No II

Flumazenil Benzodiazepine toxicity Yes Yes Yes IIIGlucagon hydrochloride

§�-Blocker, calcium channel

blockerYes Yes Yes III

Methylene blue Methemoglobinemia Yes Yes Yes IINaloxone hydrochloride Opioid and opiate drugs Yes Yes Yes IOctreotide acetate

§Sulfonylurea-induced

hypoglycemiaYes Yes No II

Physostigmine salicylate Anticholinergic syndrome Yes Yes Yes IIPotassium iodide Thyroid radioiodine protection Yes Yes No IIIPralidoxime chloride Organophosphorus insecticide

poisoningYes Yes NC II

Pyridoxine hydrochloride Isoniazid, hydrazine andderivatives

Yes Yes Yes III

Prussian blue Thallium/radiocesium NC NC NC IISodium bicarbonate

§Sodium channel blocking drugs,

urine or serum alkalizationYes Yes Yes II

IV, Intravenous; NA, do not apply because panel did not recommend stocking; EDTA, ethylene diamine tetraacetic acid; DTPA, diethylene triamine pentaacetic acid;NC, panel could not reach consensus.Cyanide antidote kit: conventional kit composed of amyl nitrite, sodium nitrite, and sodium thiosulfate. Class of evidence: Class I: good-quality randomized and blindedclinical trials and good-quality systematic reviews of good-quality randomized trials; class II: prospective, nonrandomized, or nonblinded clinical trials, cohort or well-designed case-control studies, good-quality observational or volunteer studies; class III: retrospective case series, case studies, relevant expert opinions, or animalstudies.16

*In most hospitals, immediately availability means that the antidote should be stocked in the ED.†Class of evidence was defined as the highest level of evidence observed.‡Preferred agent.

§Indication listed in package label does not include its antidotal use.

390 Annals of Emergency Medicine Volume , . : September

Dart et al Antidote Stocking Guidelines

Table 3. Amount of antidote typically needed to treat 1 patient weighing 100-kg.

Antidote

StockingRecommendation

Drug AWP for8 Hours, $

Notes and Considerations for Hazard VulnerabilityAssessment (HVA)8 h* 24 h*

Acetylcysteine 28 g 56 g IV: 800.30 Note: This recommendation applies to stocking of either oral or intravenous productsPO: 214.20 Note: Administer intravenously for hepatic failure

Antivenin (Crotalidae)polyvalent

30 Vials 30 Vials 36,705 Note: Product has been discontinued by manufacturer; some supplies remain

HVA: Geographic/endemic areas, history/experience with exotic bites, considersimultaneous bite victims (Table 4)

Crotalidae PolyvalentImmune Fab, ovine

12 Vials 18 Vials 18,858 Note: Less common acute and delayed antivenom reactions, faster mixing, use inequine serum hypersensitivity

HVA: Geographic/endemic areas, history/experience with exotic bites, considersimultaneous bite victims (Table 4)

Antivenin (Latrodectusmactans)

1 Vial 1 Vial 31.10 HVA: Geographic/endemic areas (Table 4)

Antivenin (Micrurusfulvius)

5 Vials 10 Vials 8,568.00 Note: Product has been discontinued by manufacturer; some supplies remaining.Antivenoms from Mexico or Costa Rica are likely effective. Contact regional poisoncenter to locate antivenom sources. HVA: Geographic/endemic areas (Table 4)

Atropine sulfate 45 mg 165 mg 140.62Botulism antitoxin,

equine (A, B)NA NA NA Note: Contact your state health department to assist with procurement from the

Centers for Disease Control and PreventionBotulism immune

globulin (BabyBIG)NA NA NA Information from Infant Botulism Treatment and Prevention Program, Telephone: 510-

231-7600, http://www.infantbotulism.org/physician/obtain.phpCalcium chloride 10 g 10 g 15.00 Note: do not administer subcutaneously. Should be administered by central venous IV

route, if possibleCalcium gluconate 30 g 30 g 26.70 Note: May be given by IV, SQ routes

Both calcium gluconate and calcium chloride should be availableCalcium disodium EDTA 0.75 g 2.25 g 58.03Calcium DTPA 1 g 1 g 70.00 HVA: receiving hospital for research laboratory (Table 4)Cyanide antidote kit 1 kit 1 kit 274.56 HVA: Industry, history, local conditions, community planning, facility service area (Table 4)

Note: Nitrites cause methemoglobinemia and can impair oxygen delivery; should notbe used in smoke inhalation patients with carbon monoxide poisoning; sodiumthiosulfate may be used but evidence limited, and may cause hypotension

Hydroxocobalaminhydrochloride

10 g 10 g 812.50 Note: Can be used safely in patients with smoke inhalation. Red color of drug causeslaboratory test interference, red discoloration of skin and urine

HVA: Industry, history, local conditions, community planning, facility service area (Table 4)Deferoxamine mesylate 12 g 36 g 417.18Digoxin Immune Fab 15 Vials 15 Vials 600.00Dimercaprol 500 mg 1.5 g 197.74Ethanol 180 g 360 g 76.56 Note: Ethanol is an effective antidote. It is only available as 95% concentration and

requires compounding at use. Loading dose, maintenance infusion, and frequentdose adjustments required. Medication errors are common

Fomepizole 1.5 g 4.5 g 1,364.85 Note: Fomepizole preferred for simplicity of use, lack of need for compounding inpharmacy, reduction in medication errors, potential for avoiding hemodialysis inselected patients, and anticipated safety in children

Flumazenil 6 mg 12 mg 41.26 Note: Primary use is for iatrogenic oversedation. Risks may outweigh benefits inpatients with mixed/unknown overdose, chronic benzodiazepine use, seizuredisorders, head injury

Glucagon hydrochloride 90 mg 250 mg 7,875.00Methylene blue 400 mg 600 mg 40.72Naloxone hydrochloride 20 mg 40 mg 131.50Octreotide acetate 75 �g 225 �g 24.08Physostigmine salicylate 4 mg 4 mg 9.72Potassium iodide 130 mg 130 mg 12.10Pralidoxime chloride 7 g 18 g 758.66Pyridoxine hydrochloride 8 g 24 g 899.20 HVA: Industry, history, endemic conditions, community planning, facility service area (Table 4)Prussian blue NA NA NASodium bicarbonate 63 g 84 g 13.95

AWP, Average wholesale price.Cyanide antidote kit: conventional kit composed of amyl nitrite, sodium nitrite, and sodium thiosulfate. AWP is an estimate produced by commercial vendors to repre-sent the average price at which wholesalers sell drugs to physicians, pharmacies, and other purchasers. These minimum stocking recommendations are made to ap-ply to all hospitals; any facilities with a Hazard Vulnerability Assessment indicating greater or lesser need should stock appropriately.*Facilities should plan for a minimum of 8 hours unless they have mechanisms for more rapid resupply or transfer already in place. Facilities should plan for 24 hours

if they will maintain patients for longer periods or will provide definitive care. Caution: 24-hour amount may not be sufficient for the entire treatment course.

Volume , . : September Annals of Emergency Medicine 391

Antidote Stocking Guidelines Dart et al

countries.2-9 However, it is difficult for hospitals to address thissituation because widely accepted guidelines for antidote stockinghave not emerged, although certain regional guidelines have beenpromulgated.11,20-22 National guidelines are difficult to producebecause of the heterogeneity of hospital organization andmanagement, as well as the diversity of service area. The expertpanel therefore concluded that a mechanism allowingcustomization of stocking for each hospital should be used.

To allow customized application of these guidelines, thepanel developed the concept of an antidote hazard vulnerabilityassessment, an adaptation of the Hazard VulnerabilityAssessment required in the United States for accreditation ofhospitals by TJC. As defined by TJC, a Hazard VulnerabilityAssessment is the identification of potential emergencies and thedirect and indirect effects these emergencies may have on thehospital’s operations and the demand for its services.23 Thisprocess is already required of hospitals that are accredited byTJC and provides a useful framework to assess contingenciespresented by poisoned patients. All hospitals should perform anantidote hazard vulnerability analysis. Myriad variables such assize, administrative structure, specialty, and local characteristics

Table 4. Hazard vulnerability assessment for emergency antido

Factor Principle

Pharmaceutical productsused as therapeuticagents

Agents that are widely available shouldgenerally have the antidote stockedbecause important geographicdifferences are not anticipated

Characteristics ofhospital catchmentarea

Industries, practices, activities, andindigenous fauna indicate potentialneed for antidote

Referral patterns Many hospitals accept referrals fromremote areas. These should beincluded in risk assessment

History or experience ofuse

Some modes of suicide or abusebecome locally prevalent without aspecific industry being present

Anticipated volume ofuse

Depending on characteristics of area,more than 1 victim of a poisoningmay be anticipated

Anticipated time torestocking or resupplyof antidote

Time to restocking varies greatlyamong hospitals

may affect institutional antidote needs. The hazard analysis

392 Annals of Emergency Medicine

approach holds promise in facilitating the appropriateassessment of antidote stocking needs.

The hazard assessment concept requires a hospital toformally analyze the need for an antidote and the amount ofeach antidote needed for their facility. Prioritization of risks isbased on available objective data (hospital services andutilization, demographic information, local industrial uses,availability of antidote at neighboring facilities, and chemicaltransportation routes, among other factors) that requireinteraction with appropriate businesses and manufacturers, aswell as local, state, and federal agencies. Table 4 providespotential variables that should be considered in this hazardassessment. A hospital should use the hazard assessment processto determine the treatment period for which antidote stockingshould occur. Some hospitals may exist in an environmentmaking stabilization and referral of a patient simple and rapid.Other hospitals may be subject to serious transportationdifficulties and extreme weather conditions. The process ofhazard assessment should include all stakeholders: for example,pharmacy, emergency medicine, clinical toxicology, ICU, riskmanagement, nursing, pharmacy and therapeutic committee,

Example

etaminophenticholinergic agents

enzodiazepinesapsoneigoxinononiaziddocainepioid analgesicsulfonylurea hypoglycemic agentsdustries generating or using cyanide, heavy metals, hydrogen fluoride,organophosphorus chemicals, radionuclides, thallium

hemical transportation routesdigenous fauna and flora (snakes, spiders, plants)ricultural practices (organophosphate insecticides, cyanide baits,mining)ansfers to urban hospital from agricultural areaseferral from mining region

pularity of cyanide or other specific agents as a suicide agentateur snake keepers in area

esidential or commercial fires (older buildings, lack of fire alarms, etc)ultiple casualty incidents (eg, smoke inhalation involving treatmentwith cyanide antidotes)digenous crotaline snakebite in areas with frequent occurrences suchas the southeastern or southwestern United States

ospitals that stabilize and refer patients to other institutions shouldstock for the anticipated period

ospitals that provide tertiary or definitive treatment should stock foranticipated duration of illness or until restocking from anotherhospital or distributor can occurme to restocking varies by antidote. Some may have prolongedperiods before restocking can occur

tes.

AcAnBDDIrIsLiOSIn

CInAg

TrR

PoAmRM

In

H

H

Ti

hospital preparedness committee, and hospital administration.

Volume , . : September

Dart et al Antidote Stocking Guidelines

The regional poison center is an important resource to includein the assessment process.

Some hospitals may forgo stocking of some antidotes,optimistically concluding that antidotes can be obtained quicklyfrom neighboring facilities in case of urgent need. However, theexperience of the expert panel indicates that delays are oftenencountered during the transfer of antidotes from one hospitalto another, even between neighboring hospitals or hospitalsunder the same management, thereby compromising patientcare. Delays can arise from the lack of a dedicated system tofacilitate transfer, the infrequent and unplanned nature of theserequests, and difficulties prioritizing the delivery of a medicationto another facility over urgent internal hospital orders.Infrequently used antidotes may be difficult to find during anemergency, even within the same facility.22 To address thisissue, some facilities have created charts listing antidotes andtheir location within that hospital, some facilities have created aspecial area in the pharmacy specifically for the stocking ofantidotes, whereas other facilities have created a poisoning cartsimilar to a cardiac arrest cart.22 It is recommended that eachfacility ensure that the place and the amount of each antidotestocked are known and accessible to appropriate hospitalpersonnel within the period designated by the antidote expertpanel.

These recommendations are not intended to create astandard of care. The recommendations are specifically createdfor consideration by hospitals in preparing for clinical demandsin their facility. Furthermore, antidote use will change asmedical practice evolves and the characteristics of poisoning andoverdose change. In addition, each hospital is faced with uniquesocial, political, and geographic challenges that may alter therecommended amount of antidote to stock.

The cost of a specific antidote is considered an importantfactor in hospital pharmacy purchasing decisions. Although thepurchase price of some antidotes can appear expensive, theoverall effect on the pharmacy expense budget is smaller than itmay appear because they are infrequently used and can at timesbe returned on expiration if unopened.12 According to averagewholesale price, the maximum total cost of all antidotesrecommended by the panel to treat the minimum number ofpatients would be approximately $70,000 to stock antidotes foran 8-hour treatment period and $90,000 for a 24-hour period.The primary components of this cost are rattlesnake antivenom,digoxin Fab, and glucagon. Finally, many institutions actuallystock more of an antidote than is appropriate.11 Strategies tominimize costs include reducing inappropriate use and wastefuloverstocking, regional stock rotation, and sharing multivialpacks between facilities.

The stocking of antidotes has remained a persistent concernfor at least 25 years. The use of the recommendations of theconsensus panel, combined with a hospital antidote hazardvulnerability assessment, will allow a hospital to prepare

appropriately for the treatment of poisoned patients.

Volume , . : September

Supervising editors: Lewis S. Nelson, MD; Michael L. Callaham, MD

Dr. Nelson and Dr. Callaham were the supervising editors onthis article. Dr. Dart did not participate in the editorial reviewor decision to publish this article.

Panel decisions: Panelists without a conflict of interest wereeligible to vote on any issue. Panelists with conflicts wereprohibited from voting on the drugs involved in the competinginterest. The chair was nonvoting and disclosed that DenverHealth’s Rocky Mountain Poison and Drug Center is anonprofit governmental facility that provides poison and druginformation, and research and consulting services to variousentities under contract.

Funding and support: By Annals policy, all authors are requiredto disclose any and all commercial, financial, and otherrelationships in any way related to the subject of this article,that might create any potential conflict of interest. See theManuscript Submission Agreement in this issue for examplesof specific conflicts covered by this statement. Completedisclosures can be found in Appendix E1, available online athttp://www.annemergmed.com.

Publication dates: Received for publication July 4, 2008.Revisions received September 2, 2008, and January 10,2009. Accepted for publication January 16, 2009. Availableonline May 5, 2009.

Presented at XXVIII International Congress of the EuropeanAssociation of Poisons Centres and Clinical Toxicologists, May2008, Seville, Spain.

Reprints not available from the authors.

Address for correspondence: Richard C. Dart, MD, PhD, 777Bannock Street, Mailcode 0180, Denver, CO 80204; E-mail:rdart@rmpdc.org.

REFERENCES1. Bronstein AC, Spyker DA, Cantilena LR, et al. 2006 Annual report

of the American Association of Poison Control Centers’ NationalPoison Data System (NPDS). Clin Toxicol. 2007;45:815-917.

2. Howland MA, Weisman R, Sauter D, et al. Nonavailability ofpoison antidotes. N Engl J Med. 1986;314:927-928.

3. Chyka PA, Conner HG. Availability of antidotes in rural and urbanhospitals in Tennessee. Am J Hosp Pharm. 1994;51:1346-1348.

4. Dart RC, Stark Y, Fulton B, et al. Insufficient stocking of poisoningantidotes in hospital emergency departments. JAMA. 1996;276:1508-1510.

5. Nogue S, Soy D, Munne P, et al. Antidotes: availability, use andcost in hospital and extra-hospital emergency services ofCatalonia (Spain). Arch Toxicol. 1997;19:299-304.

6. Ong HC, Yang C-C, Deng J-F. Inadequate stocking of antidotes inTaiwan: is it a serious problem? Clin Toxicol. 2000;38:21-28.

7. Juurlink DN, McGuigan MA, Paton TW, et al. Availability ofantidotes at acute care hospitals in Ontario. CMAJ. 2001;165:27-30.

8. Gorman SK, Zed PJ, Purssell RA, et al. Antidote stocking in BritishColumbia hospitals. CJEM. 2003;5:12-17.

9. Hruby K. Dostupnost antidote v nemocnicnich lekarnach ceskerepubliky. Ces Slov Farm. 2003;52:231-240.

10. Dart RC, Goldfrank LR, Chyka PA, et al. Combined evidence-based

literature analysis and consensus guidelines for stocking of

Annals of Emergency Medicine 393

Antidote Stocking Guidelines Dart et al

emergency antidotes in the United States. Ann Emerg Med.2000;36:126-132.

11. Bailey B, Bussieres J-F, Dumont M. Availability of antidotes inQuebec hospitals before and after dissemination of guidelines.Am J Health Syst Pharm. 2003;60:2345-2349.

12. Sivilotti MLA, Eisen JS, Lee JS, et al. Can emergencydepartments not afford to carry essential antidotes? CJEM. 2002;4:23-33.

13. Joint Commission on Accreditation of Healthcare Organizations.Comprehensive Accreditation Manual for Hospitals. OakbrookTerrace, IL: Joint Commission on Accreditation of HealthcareOrganizations; 2007:111.

14. California Department of Public Health. Hospital administrationpenalties by county. Available at: http://www.cdph.ca.gov/certlic/facilities/Documents/HospitalAdministrativePenalties-2567Forms-LNC/2567OliveViewUCLAMedical-Sylmar-EventE-EW2J11.pdf.Accessed May 22, 2008.

15. Flomenbaum NE, ed. Goldfrank’s Toxicological Emergencies. 8thed. New York, NY: McGraw-Hill; 2006.

16. Dart RC et al. Medical Toxicology. 3rd ed. Philadelphia, PA:Lippincott Williams & Wilkins; 2004.

17. Wolf SJ, Heard K, Sloan EP, et al. Clinical policy: critical issues in

the management of patients presenting to the emergency

394 Annals of Emergency Medicine

department with acetaminophen overdose. Ann Emerg Med.2007;50:292-313.

18. McDowell MA, Fryar CD, Hirsch R, et al. AnthropometricReference Data for Children and Adults: US Population, 1999-2002. Advance Data From Vital And Health Statistics; No. 361.Hyattsville, MD: National Center for Health Statistics; 2005.

19. Berberet BR, Burda BR, Lodolce AE. Unavailability of 10% alcoholin 5% dextrose injection. Am J Health Syst Pharm. 2005;62:2344-2345.

20. Berndt E. Posted list of emergency drugs and antidotes. Am JHosp Pharm. 1994;51:2602.

21. Pettit HE, McKinney PE, Achusim LE, et al. Toxicology cart forstocking sufficient supplies of poisoning antidotes. Am J HealthSyst Pharm. 1999;56:2537-2539.

22. Ontario Poison Centre guidelines for stocking of emergencyantidotes. Available at: http://www.ontariopoisoncentre.com/ontariopoisoncentre/custom/antidotes2006.pdf. Accessed May4, 2008.

23. The Joint Commission on Accreditation of Healthcare Organization.2009 Hospital Accreditation Standards. Oakbrook, IL: The JointCommission; 2008. Available at: http://www.jointcommission.org/NR/rdonlyres/F42AF828-7248-48C0-B4E6-BA18E719A87C/0/06_

hap_accred_stds.pdf. Accessed May 6, 2008.

Volume , . : September

Appendix E1. Disclosures.Funding for this research was provided by 8 pharmaceutical com-panies (Fougera, Dey, Protherics, Rare Disease Therapeutics,

Heyl, Cumberland Pharmaceuticals, Heyltex, and Jazz Pharma-

Volume , . : September

ceuticals) to the Denver Health and Hospital Authority. Thesponsors had no input into the design, definitions, panel proceed-ings, analysis of the results, or drafting of the article.

Dr.Heard received support fromNIHGrant (1K08DA020573-01).

Panelist OrganizationFinancialInterests Clinical Grants Consulting

Potential AntidoteInterests

Borron University of Texas HealthSciences Center at SanAntonio

None EMD Pharmaceuticals, Dey LP(Hydroxocobalamin)

Merck KGA (Hydroxocobalamin) Hydroxocobalamin

Caravati Utah Poison Control Center None None None NoneCobaugh American Society of Health-

System PharmacistsResearch & EducationFoundation

None None None None

Curry Banner Good SamaritanMedical Center

None None Fougera (CroFab UnrestrictedEducation Grant forToxicology Fellowship,Lectures-No Honoraria)

Crotaline Immune Fab

Dart Rocky Mountain Poison &Drug Center-Denver Health

None See Note Below See Note Below Not Applicable - Non-VotingChair of Panel

Falk Orlando Regional HealthcareSystem

None None None None

Goldfrank New York University Schoolof Medicine

None None None None

Gorman Centers for Disease Control& Prevention

None None None None

Groft Office of Rare Disease atNational institutes ofHealth

None None None None

Heard Rocky Mountain Poison &Drug Center-Denver Health

None Fougera (CroFab Registry),Cumberland (NAC Registry),RDT (BWS Investigator),Protherics (DigiFabInvestigator)

None NAC (IV), Crotaline ImmuneFab, DigiFab, BWS AV

Miller Orange County Fire Authority/Orange County HealthcareAgency

None None None None

Olson California Poison ControlSystem

None None None None

O’Malley Albert Einstein MedicalCenter

None None None None

Schaeffer Rocky Mountain Poison &Drug Center-Denver Health

None RDT (BWS Co-Investigator),Protherics (DigiFab RetroStudy)

None BWS AV, DigiFab

Seger Tennessee Poison Center None None None NoneSeifert University of New Mexico/

New Mexico Poison & DrugInformation Center

None Protherics (Investigator CroFab),RDT (Investigator Anavip),HRSA (Strategic AV SystemGrant)

Poisindex Editorial Board Crotaline Immune Fab

Sivilotti Queen’s University None Cumberland (Unrestrictedresearch grant)

None NAC (IV)

Tomassoni Yale New Haven Health None None None NoneWise Joint Commission None None None None

Note: Denver Health’s Rocky Mountain Poison & Drug Center is a non-profit governmental facility that provides poison & drug information, research and consulting ser-vices to various entities under contract. Clients of the Rocky Mountain Poison & Drug Center include a variety of governmental, nongovernmental and commercial orga-nizations. In addition to the sponsors of the Antidote Summit, these include Eli Lilly, Abbott Laboratories, Wyeth Pharmaceuticals, Roche Pharmaceuticals, and others.

Annals of Emergency Medicine 394.e1