Epipen epidemic: suggestions for rational prescribing in childhood food allergy

J. Paediatr. Child Health (2003) 39, 372–375
EpiPen epidemic: Suggestions for rational prescribing
in childhood food allergy
Department of Allergy, Immunology and Infectious Diseases, The Children’s Hospital at Westmead, Westmead, New South Wales, Australia Abstract:
There has been a marked increase in community concerns of the risk of food induced anaphylaxis in children and a consequent increase in the provision of the self or carer injectable epinephrine (EpiPen) (CSL Ltd, Parkville, Victoria,Australia)). The Australian use of EpiPens in children under 10 years has increased by 300% over 5 years with a crude rateof EpiPen provision of 1 per 544 Australian children aged under 10 years. However, the risk of a fatal reaction to food,particularly in preschool children, is remote (in Australia, an estimated one fatality in 30 years in the under 5-year-oldpopulation and two deaths in 10 years in the entire child population). It is therefore important to provide a perspective onthe risk of death from food induced anaphylactic to parents and carers in view of the anxiety generated on this issue. Theindications for provision of an EpiPen to children are not well defined. Six risk factors, which can be considered inevaluating the risk of a life-threatening reaction (age over 5 years; a history of respiratory tract involvement with the initialor subsequent reactions; a history of asthma requiring preventer medication; peanut or tree nut sensitivity; reactions inducedby traces or small amounts of allergen; a strongly positive skin prick test) are proposed. It is suggested that the greater thenumber that are positive, the lower the threshold for provision of an EpiPen. In addition, instruction in EpiPen administrationand the provision of both a clear and simple anaphylaxis action plan and a rational perspective on the remote risk of death isjust as important as the provision of the device itself.
Key words:
anaphylaxis; epinephrine; EpiPen; food.
INTRODUCTION
period2 and Hourihane et al. found peanut allergy increasedfour-fold between generations.3 Allowing a time span of Community concerns concerning the risk of food-induced 30 years between generations this would indicate a 66% anaphylaxis and death in children have multiplied in recent increase over 5 years while EpiPen Jr. usage has increased by years. Statements by lay and consumer groups, such as the 300% over the same time period. In Australia, in 2000, there following ‘thousands of families with young children are were approximately 2.6 million children under 10 years of age.
forced to live with the possibility that everyday foods may be Assuming all EpiPen Jr. were prescribed for children under contaminated by a known allergen which could kill in min- 10 years of age, this gives a crude rate of EpiPen provision of utes’1 reflect the anxiety which parents feel. This anxiety, 1 per 544 children, which is in excess of estimates for children together with the incorrect perception that allergic reactions in some other countries: 1/1600 in UK,4 and 1/5800 in France5 invariably increase in severity, which is common among both but less than a remarkable 1/84 of a population of 280 000 lay people and the medical profession, has generated a potent children in the province of Manitoba, Canada.6 While the rate pressure for action by the medical profession to avert these refers to the number of units per child in the community and perceived threats. A combination of increased community not the number of children with EpiPens, which may be less concern and an increase in awareness of anaphylaxis is the if the practice of prescription of more than one unit per child likely cause of the rapid increase of prescriptions for self or is widespread, it does accord with the population survey of carer administered epinephrine EpiPen (EpiPen; CSL Ltd, provision of some form of adrenaline of 1/417 in a sample Parkville, Victoria, Australia) in recent years (Table 1). In of 4173 South Australian children.7 The appropriate rate of Australia, all EpiPens are supplied by CSL Ltd. and, thus their provision of EpiPens in a given population is yet to be sales figures reflect the Australian usage. Over 5 years, the total determined. In these calculations, it has been assumed that all number of units provided to retail pharmacies and hospitals has EpiPens are provided for food anaphylaxis. While personal increased by 193% and the provision of EpiPen junior (EpiPen experience indicates that the majority of EpiPens in children Jr.), which is recommended for children under 30 kg (i.e. under (> 95%) are provided for food allergy, these figures will about 10 years of age), a dramatic 300%. Although allergic overestimate the rate to the extent that EpiPens are also diseases are increasing, it does not appear that the rapid prescribed for other conditions such as bee sting anaphylaxis.
increase in EpiPen usage is primarily due to this factor.
Concurrent with the marked increase in prescriptions in Sampson observed in his patients that peanut sensitization (a Western communities has been controversy in the medical positive skin prick test) had increased by 55% over a 10-year literature concerning benefits and risks with some authorities Correspondence: Professor A Kemp, Department of Allergy, Immunology and Infectious Diseases, The Children’s Hospital at Westmead, Westmead, NSW 2145, Australia. Fax: +61 2 9845 3421; email: andrewk5@wch.edu.au Accepted for publication 11 November 2002.
Number of EpiPen units provided to Australian retail history of anaphylaxis or serious reaction and the risk of pharmacies and hospitals (annual totals to March each year are shown) another reaction is substantial in the judgment of the clinician,an epinephrine kit should be prescribed with clear instructions regarding its use’.13 This recommendation invites the clinicianto consider what is a substantial risk and, in addition, implies that adrenaline is not necessarily prescribed for any level of risk. A recent review of deaths in the USA concluded, ‘patients at risk for food-induced anaphylaxis (i.e. those with previous reactions involving the airway or those with asthma and foodallergies) must be educated to recognize the early signs ofanaphylaxis and be given and trained in the use of self-injectable epinephrine’.11 This conclusion reflects the fact thatthe greater majority of deaths occur in patients with significant expressing the opinion that EpiPens are over prescribed8 and asthma and also suggests that the severity of the presenting challenge to this by others.9 The response to the remote risk of reaction (i.e. previous reaction involving the airway) might be death in a child from food anaphylaxis is essentially about risk utilized as an indicator of the risk of a subsequent severe management. A perspective on this issue is needed for paedi- reaction. However, in another publication, these authors expand atricians to inform their patients, and indications for rational the potential population by stating ‘young children with peanut prescribing are urgently required. This paper will attempt to and/or tree nut reactions should be considered at risk for more severe reactions and should be provided with emergency medi-cations (epinephrine)’.14 In a recent review, Sampson stated‘All patients with peanut allergy should be given a written THE RISK OF DEATH IS LESS IN CHILDREN UNDER
emergency plan and adequate doses of liquid diphenhydramine and self-injectable epinephrine for use in case they accidentallyingest peanuts’.15 A study in the UK identified eight deaths in children (pop-ulation of 13 million children) over a period of 10 years fromfood anaphylaxis.10 Another survey in the USA, between 1994 ESTIMATE OF COST TO AUSTRALIAN COMMUNITY
and 1999, identified 32 fatalities, of which nine were under16 years of age, due to food-induced anaphylaxis from a It is possible to make some estimate of the potential cost to the national registry.11 If one extrapolated the UK figures to the Australian community for the provision of EpiPens to children Australian childhood population, there would be one death in under 16 years with peanut or tree-nut allergy if the recommen- 30 years in the under 5-year-old population and two deaths dations of Sampson that all such patients should be provided in 10 years in the entire child population. The great majority with an EpiPen are followed.15 Adopting the US estimate that of deaths occur in children aged 5 years and older and adults, 1.1% of the population has peanut or tree-nut allergy,15 (which 7/8 (88%) in the UK study and 30/32 (94%) in the USA. In is likely to be an underestimate for the under 16 year age group contrast, food-induced immediate hypersensitivity reactions as peanut allergy is more common in children than adults) there occur most commonly in preschool children to milk, egg and would be 46 200 children under 16 years with peanut or tree- peanuts; frequently, in the case of egg and milk, and less nut allergy in Australia. The current recommended retail cost commonly with peanut they resolve by 5 years of age. How- for an EpiPen is $A140.00. To ensure appropriate 24 h cover- ever, as it is those children with food allergy who are the ones age, many children are prescribed two units, one for home and at risk, it is necessary to consider the risk of death from food one for kindergarten/school. With a shelf life of 15 months, this anaphylaxis in this subpopulation. Up to 5% of the childhood provides an annual cost of $10.35 million dollars if each child population may have food allergies.12 Assuming all the deaths has two units. From the figures provided above, the estimated occurred in this group, there would be 0.3 deaths in Australia in death rate for children under 16 years in Australia is one child 10 years in the estimated 65 000 food allergic children under every 5 years, and assuming all deaths in childhood were due to 5 years of age. This may be compared with the death rate from nut allergy and were preventable by provision of an EpiPen this vaccine preventable meningococcal infection in the same gives a cost of $51.7 million dollars per life saved. In attempt- 65 000 children of 1.2 deaths in 10 years (15 deaths/year from ing to evaluate the costs and benefits of EpiPen usage, there meningococcal disease in Australian children, half occur in may be other benefits unrelated to prevention of death such as under 5-year-old age group and one-third preventable by a reduction in morbidity and increased quality of life, however, immunization). Thus, a food allergic child aged under 5 years there is insufficient data to enable a cost analysis to be made.
may be four times more likely to die from a preventablemeningococcal infection than from food anaphylaxis. Thisexample does not suggest that appropriate emergency measures INDICATIONS FOR PRESCRIPTION OF EPIPEN
should not be prescribed where indicated but does indicate thatperspective needs to be applied in the assessment of risks to As the prescription of an EpiPen is primarily concerned with health and choices concerning expenditures.
risk management in attempting to develop some rational guide-lines, it seems appropriate to ask whether there are factorswhich might point to the likelihood of developing a severe life- GUIDELINES FOR EPIPEN PRESCRIPTION ARE
threatening reaction. I suggest that there are six, which may be considered.
1. Age over 5 years.
Currently there are no clear guidelines on which children 2. A history of respiratory tract involvement with the initial or should be prescribed an EpiPen. In 1994, the American Academy of Allergy and Immunology stated ‘If there is a 3. A history of asthma requiring preventer medication.
Another factor, which may help, is an assessment of the 5. Reactions induced by traces or small amounts of allergen.
amount of food allergen ingested and the severity of reaction 6. A strongly positive skin prick test.
induced. Although this has not been studied in detail, many Few would disagree to provision of an EpiPen to a child, of authorities believe that the severity of an adverse reaction bears any age with peanut sensitivity, who has developed a previous some relationship to the amount of allergen ingested, as evi- reaction involving the respiratory tract. In contrast, the indica- denced by recommendations to avoid challenges or commence tions are less clear for a preschool child with a generalized with lower doses in severely food allergic subjects.20–22 That urticarial reaction, without respiratory tract involvement, fol- the amount of allergen ingested is related to the severity of a lowing ingestion of 50 mL of cow’s milk and a 5-mm skin reaction is also evidenced by the common practice of perform- ing food challenges with gradually increasing doses of food in The great majority of fatalities are recorded in children are order to reduce the risk of a severe reaction.21–23 In adults, over 5 years of age, despite the fact that food allergic reactions repeat challenges of peanut protein on two occasions have been are more common in preschool children and frequently lessen remarkably consistent in the threshold doses required to elicit with time. In the survey of deaths in the USA over 6 years, clinical reactions in peanut sensitive subjects.20 Moneret- there were two deaths in children less than 5 years; one to Vautrin et al. utilized the dose eliciting symptoms as one factor Brazil nut, which occurred on the first exposure and thus would in the decision to prescribe an EpiPen.5 Thus, a child who not have been preventable, and a second to cows milk.11 In develops mild generalized urticaria following ingestion of half Sweden, a prospective survey conducted over 4 years identified an egg is likely to be in a different risk category to one with a six fatal reactions in children, two to peanut and four to soy, all respiratory reaction following exposure to a small amount of of whom were older than 8 years.16 In the UK, the single fatality in a child under 5 years was due to egg white.10 Thus, An important and unresolved issue is whether the size of the in a total of 14 years observation, in three countries, of the 46 skin prick test bears a relation to the likelihood of a severe fatal reactions recorded, only three (7%) occurred in children reaction, and should be taken into account when deciding to prescribe an EpiPen. In a large group of both adults and It has been suggested that the severity of the initial reaction children, overall skin prick test size did correlate with severity be taken into account when prescribing an EpiPen.5,9,11 Bock (P = 0.04); however, the substantial overlap between groups et al. stated that a previous history of airway reactions was an lessens the predictive power in any individual case.17 In chil- indication for prescription of EpiPen.11 An unresolved issue is dren, the size of the skin prick test was related to the occurrence how predictive of subsequent severity is the initial reaction.
of a clinical reaction on formal food challenge. An Australian Whilst it appears clear that a severe reaction can follow a mild study of Sporik et al. demonstrated that a wheal size 8 mm reaction there is evidence to indicate that the severity of the or greater was associated with a 100% (28/28) reactivity on initial reaction provides some prediction of the nature of a peanut challenge while a wheal of 2–4 mm was associated with subsequent reaction. Hourihane et al. surveyed over 600 aller- only 47% (8/17) reactivity.22 Pucar et al. performed a peanut gic reactions in adults and children to peanuts. Of the patients challenge in children with a positive skin prick test and 46% whose first reaction was severe (wheeze, cyanosis, collapse or (13/28) of those with a skin prick test wheal 7 mm or greater faint), the most recent reaction was also severe in 78% of cases reacted clinically, while only 14% (5/36) of those with a wheal (173/223) whilst in those cases with a mild reaction the most size less than 7 mm reacted. Two of the 18 children with a recent reaction was severe in 32% of cases (34/107).17 Ewan positive challenge developed wheeze.21 However, children with and Clark found that of 15 patients with a mild initial reaction a definite history and a 5 mm or greater skin prick test reaction who had a further reaction, three (20%) had a more severe were not challenged. As food challenges are generally per- reaction in which some form of adrenaline (inhaled 2, formed with a progressive increase in dose of food and cease injected 1) was used.18 Sicherer et al., in a study of peanut once minor symptoms develop, the results obtained by a allergic reactions in children, found ‘On average, symptoms controlled challenge may not reflect the severity of the reaction after accidental exposure were generally similar to those at on exposure to larger amounts of allergen. From this data it is initial exposure’.19 In this study, approximately half the difficult to make recommendations, some may choose to ignore reactions involved the respiratory tract.
the size of the skin prick test, however, as challenge studies The presence of asthma increases the risk of death10,11 or a indicate that clinical reactivity bears some relationship to the severe reaction.14 This observation has lead to the recommen- skin prick test wheal size, it seems reasonable to factor this into dation that those subjects with both asthma and a food allergy should be prescribed an EpiPen.11 In the paediatric context, itseems reasonable to apply this to those children with asthma ofsufficient severity to require preventer medication.
SENSITIZATION WITHOUT PRIOR FOOD
It is clear that the great majority of deaths are associated INGESTION
with nut, and, in particular, peanut sensitivity. Combining themortality results from USA, UK and Sweden, 34/46 deaths A common issue in preschool children is the identification of were associated with peanut or tree-nut reactions.10,11,16 Deaths peanut sensitivity by skin prick testing in children who have from egg and cows milk sensitivity are much less common never knowingly ingested peanut but have reacted to another despite being among the most common food allergies detected.
food (usually milk or egg). In this situation, the risk of a severe Of the three deaths in the under 5 years age group, one occurred anaphylactic reaction on exposure to peanut is unknown. Pucar to milk and one to egg. In Sweden, there were four deaths due et al. found that, in those who have never knowingly been to soy anaphylaxis in peanut sensitive subjects.16 There are exposed to peanut but had a positive skin prick test, there was a immunological cross reactivities between soy and peanut, and 31% (5/16) prevalence of clinical reaction to peanut challenge both are members of the same botanical family. The specific and 50% (4/8) in those with a wheal size greater than 7 mm.21 properties of peanuts that lead to severe reactions are unknown; In view of this, a case can be made for formal peanut challenge however, they may relate to properties of the three major in this group to determine the clinical features of a reaction just allergenic proteins in peanuts Ara h 1, 2, 3.15 prior to school entry, and provision of an EpiPen to those with a generalized cutaneous or more severe reactions following 5 Moneret-Vautrin DA, Kanny G, Morisset M et al. Food anaphy- challenge. An alternative policy would be to provide EpiPens laxis in schools: evaluation of the management plan and the effi- to all children with a prick test wheal size greater than a ciency of the emergency kit. Allergy 2001; 56: 1071–6.
predetermined size (e.g. > 7 mm); however, on the evidence of 6 Simons FE, Peterson S, Black CD. Epinephrine dispensing for the out-of-hospital treatment of anaphylaxis in infants and children: a Pucar et al.,21 this would be unnecessary in 50% of cases and, population-based study. Ann. Allergy Asthma Immunol. 2001; 86:
in addition, perpetuate a significant amount of unwarranted parental anxiety. It has been suggested that ‘no allergist would 7 Boros CA, Kay D, Gold MS. Parent reported allergy and anaphy- prescribe an epinephrine kit on the basis of a positive skin prick laxis in 4173 South Australian children. J. Paediatr. Child Health test in the absence of a significant history or formal challenge’.9 2000; 36: 36–40.
8 Unsworth DJ. Adrenaline syringes are vastly over prescribed.
Arch. Dis. Child. 2001; 84: 410–1.
CONCLUSIONS
9 Hourihane J. Controversies in paediatrics? Arch. Dis. Child. 2001; It can be seen that there are no clear-cut guidelines for the 10 Macdougall CF, Cant AJ, Colver AF. How dangerous is food provision of an EpiPen to a child with food allergy. This, in allergy in childhood? The incidence of severe and fatal allergic part, reflects the difficulty in predicting those children most reactions across the UK and Ireland. Arch. Dis. Child. 2002; 86:
236–9.
likely to be at risk and in deciding the limits of an acceptable 11 Bock SA, Munoz-Furlong A, Sampson HA. Fatalities due to risk. I have listed six factors that can be taken into consider- anaphylactic reactions to foods. J. Allergy Clin. Immunol. 2001; ation. It would be generally agreed that a child over 5 years of 107: 191–3.
age, with a history of respiratory involvement following expo- 12 Sampson HA. Food allergy. Part 1: immunopathogenesis and sure to peanut, should be provided with an EpiPen; however, clinical disorders. J. Allergy Clin. Immunol. 1999; 103: 717–28.
many cases are not as clear-cut. I suggest that each factor be 13 AAAI Board of Directors. The use of epinephrine in the treatment considered and, the greater the number that are positive, the of anaphylaxis. J. Allergy Clin. Immunol. 1994; 94: 666–8.
lower the threshold for prescribing an EpiPen. Each of these 14 Sicherer SH, Furlong TJ, Munoz-Furlong A, Burks AW, factors will need to be assessed by the practitioner and weighed Sampson HA. A voluntary registry for peanut and tree nut allergy:characteristics of the first 5149 registrants. J. Allergy Clin. in the light of the parental wishes and environmental circum- Immunol. 2001; 108: 128–32.
stances. It must be emphasized that the prescription of an 15 Sampson HA. Clinical practice. Peanut allergy. N. Engl. J. Med. EpiPen alone is not a satisfactory response to the risk of food 2002; 346: 1294–9.
anaphylaxis. This is highlighted by the Australian evidence that 16 Foucard T, Malmheden Yman I. A study on severe food reactions only 29% of EpiPens prescribed were used appropriately in a in Sweden – is soy protein an underestimated cause of food subsequent anaphylactic reaction.24 Thus, instruction in EpiPen anaphylaxis? Allergy 1999; 54: 261–5.
administration and the provision of both a clear and simple 17 Hourihane JO, Kilburn SA, Dean P, Warner JO. Clinical charac- anaphylaxis action plan and a rational perspective on the teristics of peanut allergy. Clin. Exp. Allergy 1997; 27: 634–9.
remote risk of death is just as important as the provision of 18 Ewan PW, Clark AT. Long-term prospective observational study of patients with peanut and nut allergy after participation in a man- agement plan. Lancet 2001; 357: 111–5.
19 Sicherer SH, Burks AW, Sampson HA. Clinical features of acute allergic reactions to peanut and tree nuts in children. Pediatrics ACKNOWLEDGEMENTS
1998; 102: e6.
20 Nelson HS, Lahr J, Rule R, Bock A, Leung D. Treatment of I thank Louisa Garone of CSL Ltd. (CSL Ltd, Parkville, anaphylactic sensitivity to peanuts by immunotherapy with injec- Victoria, Australia) for providing Australian figures on EpiPen tions of aqueous peanut extract. J. Allergy Clin. Immunol. 1997; 99: 744–51.
21 Pucar F, Kagan R, Lim H, Clarke AE. Peanut challenge: a ret- rospective study of 140 patients. Clin. Exp. Allergy 2001; 31:
REFERENCES
22 Sporik R, Hill DJ, Hosking CS. Specificity of allergen skin testing 1 Gowland MH. Food allergen avoidance: risk assessment for life.
in predicting positive open food challenges to milk, egg and Proc. Nutr. Soc. 2002; 61: 39–43.
peanut in children. Clin. Exp. Allergy 2000; 30: 1540–6.
2 Sampson HA. Managing peanut allergy. BMJ 1996; 312: 1050–1.
23 Taylor SL, Hefle SL, Bindslev-Jensen C et al. Factors affect- 3 Hourihane JO, Dean TP, Warner JO. Peanut allergy in relation to ing the determination of threshold doses for allergenic foods: heredity, maternal diet, and other atopic diseases. results of a how much is too much? J. Allergy Clin. Immunol. 2002; 109:
questionnaire survey, skin prick testing, and food challenges. BMJ 1996; 313: 518–21.
24 Gold MS, Sainsbury R. First aid anaphylaxis management in 4 Dobbie ARC. Provision of self injectable adrenaline for children at children who were prescribed an epinephrine autoinjector device risk of anaphylaxis: its source, frequency and appropriateness of (EpiPen). J. Allergy Clin. Immunol. 2000; 106: 171–6.
use, and effect. AMB Child Health 1998; 4: 289–94.

Source: http://not.gothic.net.au/pic/other/ebay/Epipen1.pdf