MHK Ho, LM Wong, SC Ling, KP Lee, WW Cheng, J Hui, D Lau, KL Ng, D Wong, KL Kwok, EYW Kwan, on behalf of the Clinical Audit Working Group of the Quality Assurance Subcommittee in Paediatrics

Abstract

Background: Anaphylaxis and severe systemic allergic reaction are potentially life-threatening conditions. There is a paucity of data on the management of such condition amongst Hong Kong children. Objective: This review was designed to assist health professionals to evaluate the current process of care for children admitted with anaphylaxis or severe systemic allergic reaction, to identify service gaps so that patients are appropriately investigated, treated and taught how to recognise and manage severe allergic reactions. Methods: The anaphylaxis and severe allergic reaction/angioedema for children under age of 18 were identified using ICD-9 codes 995.0, 995.1, 995.6. We performed a retrospective chart review of one hundred children. We assessed the clinical practice among all Paediatric Departments within the Hospital Authority (HA) from January 2006 to December 2007. Results: The standardised admission rates of anaphylaxis aged below 18 years was 0.5 /100,000 (CI 0.6-0.7), which was probably 5-7 fold less compared with Western countries. This territory wide survey confirmed that food allergy was the leading cause in systemic allergic reactions amongst children admitted to HA service. Drug was the second commonest cause, accounting for 24% of the cases. The causes could not be determined in one out of six cases (17%). The attempt to identify the exact aetiologies was hampered by the lack of allergy assessment in most of the units. Adrenaline auto-injector was infrequently prescribed and used in our practice. Conclusion: This study provided background information for possible implementation of improvement measures that might be needed in the management of this possible life-threatening reaction.

Introduction

Anaphylaxis is an increasing emergency in Western countries, especially in children.^1,2 Anaphylaxis is a severe and sudden allergic reaction. It can occur when a susceptible person is exposed to an allergen (such as a food or insect sting). Although death is rare, an anaphylactic reaction always requires an emergency response. Prompt treatment with injected adrenaline is required to halt progression and can be life saving. Fortunately anaphylactic reactions are usually preventable by implementing strategies for avoiding allergens. Common allergens that can trigger anaphylaxis are: foods (e.g. peanuts and other nuts, shellfish and fish; and in pre-school age children, milk and egg), insect stings (e.g. bee, wasp, jack jumper ants) medications (e.g. antibiotics, aspirin) latex (e.g. rubber gloves, balloons, swimming caps).

In western countries, the commonest cause of anaphylaxis in children is related to food allergy.^1-3 Recent local community survey indicated that parent-reported adverse food reaction is a common atopic disorder in Hong Kong pre-school children and prevalence rates are comparable to the Caucasians.⁴ This study identified that about 10% have severe reaction such as respiratory difficulties and shock. However, we have a relative paucity of clinical data on management of severe allergic reaction and anaphylaxis.

Discrepancies in the operational definitions and diagnostic criteria of anaphylaxis represent one of the most controversial issues in defining its epidemiology.¹ The lack of reliable markers of the disease hampers its diagnosis. Previous report of single centre experience indicated a lack of auto-injector of adrenaline in Hong Kong.⁵ The under-use of adrenaline is another important issue, as available overseas data demonstrates physicians' preference for steroid and antihistamines despite the current lack of evidence of their effectiveness.⁶

This study on management of anaphylaxis and severe systemic allergic reaction was designed to assist health professionals to evaluate and optimise their care of patients diagnosed with anaphylaxis to ensure that patients are appropriately investigated, treated and taught how to recognise and manage severe allergic reactions. This study has assessed clinical practice among all hospital paediatric units of Hospital Authority. The specific aims of the study included the followings: 1) To evaluate prescription patterns and procedures for auto-injectable adrenaline in paediatric units of Hospital Authority; 2) To provide information for discussion and implementation of any changes in practice that might be needed.

Methods

We obtained admissions data from Hospital Authority central computer system CDARS January 1997 to December 2007. We identified admissions for anaphylaxis and angioedema (ICD-9 codes: 995.0, 995.1, 995.6) for children under age of 18. We retrospectively reviewed 100 medical charts from January 2006 to December 2007 out of total 104 patients with the above codings. All paediatric units providing acute emergency admissions had participated in this chart review.

Operational Definition of Anaphylaxis and Selection of Disease Coding

An international widely accepted definition of anaphylaxis was adopted.

Definition of Anaphylaxis:⁷

Anaphylaxis is highly likely when ANY one of the following three criteria is fulfilled:

1. Acute onset of an illness (minutes to several hours) with involvement of the skin, mucosal tissue, or both (e.g. Generalised hives, pruritis or flushing, swollen lips-tongue-uvula) and at least one of the following:
   1. Respiratory compromise (e.g. Dyspnoea, brochospasm, stridor, hypoxia)
   2. Cardiovascular compromise (e.g. Hypotension, collapse)
2. Two or more of the following that occur rapidly after exposure to a likely allergen for that patient (minutes or several hours)
   1. Involvement of the skin or mucosal tissue (e.g. Generalised hives, itch, flushing, swelling)
   2. Respiratory compromise (e.g. Dyspnoea, bronchospasm, stridor, hypoxia)
   3. Cardiovascular compromise (e.g. Hypotension, collapse)
   4. Persistent gastrointestinal symptoms (e.g. Abdominal pain and vomiting)
3. Hypotension after exposure to known allergen for that patient (minutes or several hours):
   1. Hypotension for children is defined as systolic blood pressure <70 mmHg from 1 month to 1 year, <70 mmHg+ [2x age] from 1 to 10 years, and <90 mmHg from 11 to <18 years

The Data Form (Appendix I)

A study form covered the general concern and areas of interest related to severe allergic reaction/anaphylaxis. To ensure consistent collection of data, the data collection form was set to be simple and unambiguous. For each of the standards defined for the data collection, there should be at least one question with clear options for the answer. A pilot was conducted using the tool on a few cases that were also subsequently included in the sample. The results of the pilot indicated that the data met the study aims.

Data Handling

The study form was completed by a designated coordinator (as listed) from each unit. The form was then stored at individual units and a copy was sent to the central data manager. The patient confidentiality was of prime concern. Only a study code was recorded with individual particulars. There was however a link kept separately for future verification as deemed necessary. The data was entered into a computer with password locked.

Data Presentation and Clarification Process

There were interactive sessions for the investigators to clarify some unclear data. The process helped to recognise bias and variance and ranked problems according to priority for resolution. We also defined actions to address these problems.

Documentation and Action

The recommended actions and proposed changes in policy or procedure were discussed and documented to Coordinating Committee (COC) in Paediatrics who originally granted the approval of this study. It further granted the right for publication in order to achieve with wide circulation.

Ethics

According to the Hospital Authority research ethics guidelines, (http://www.ha.org.hk/visitor/ha_visitor_index.asp?Parent_ID=110&Content_ID=369) "Ethics approval is usually not required in a straight forward clinical audit or analysis of existing/secondary data without identifiers traceable to the subjects". This study was considered as part of an internal audit and was approved by the COC Paediatric Chairman and its respective committee members from all hospitals. The right for publication was also granted by the Committee.

Statistic Analysis

Categorical data of subgroup analysis was compared by using the 2x2 test with the Yates correction. One way ANOVA (analysis of variance) was used to comparing the difference in age among groups. All statistical analysis was performed with SAS software, version 9.1 (SAS, Inc, Cary, NC).

Results

The average calculated standardised anaphylaxis admission rate with age below 18 years due to food was 0.5/100,000 admissions (CI 0.6-0.7). The average calculated standardised admission rate with age below 18 years due to any causes was 1.5/100,000 admissions (CI 1-3) (Table 1a to 1c). The average calculated standardised angioedema admission rate with age below 18 years due to any causes was 5/100,000 admissions (CI 3-6). Generally, the rates were steady over the years.

Case Identification

There were ten admissions coded as anaphylaxis 995.0, eighty-nine coded as angioedema 995.1, and five coded as anaphylactic shock due to food 995.6. A total of 100 (96%) medical records were available for review. These included all anaphylaxis cases but 85/89 of angioedema cases. Four records coded as angioedema could not be found.

Disease Coding

We studied the coding precision by comparing the reported clinical features with a pre-defined definition of anaphylaxis. All coding for anaphylaxis were precise. Out of 15 coded as anaphylaxis, 10 had cardiovascular compromise and 5 had respiratory compromise. Whilst 85 cases coded as angioedema, none had cardiovascular compromise but (25/85, 29.4%) had symptomatic respiratory symptoms. These 25 cases should actually be coded as anaphylaxis. After revision of coding, the inter-group comparison was tabulated. Group A is after revision of coding "remained as angioedema", Group B is after revision "changed from angioedema to anaphylaxis", Group C is fulfilled the diagnosis of anaphylaxis, and was labeled as such at the outset (Table 2).

Demographic Characteristics

The male to female ratio was 55:45. The age ranged from 2 months to just under 18 years. The mean and median ages were 8.09 (SD 5.7) years and 8.21 years respectively. There was no age difference between female and male. There were 98 Chinese and 2 Non-Chinese. The incidents took place at: home 56%; restaurant 8%; school 5%; community not-specified other than medical premises 21%.

Clinical Severity

The symptoms and signs of different organs were tabulated in Table 3. Majority of them had cutaneous manifestations. A third had respiratory symptoms. Ten percent had cardiovascular compromised states. Gastrointestinal symptoms were less frequent which occurred in 8 cases. There was no single mortality after the index event. Recurrence was not evaluated. Co-factor that influenced the severity of an anaphylactic reaction (e.g. pre-existing of asthma) was not evaluated. The mean hospital stay was 2 days. Biphasic reaction, as defined by returning of symptoms after initial resolution of symptoms but without further exposure to inflicting agent,⁷ occurred in 10% of individuals.

Causes of Anaphylaxis and Severe Allergic Reactions

The presumed causes accounting for the 100 episodes were tabulated in Table 4a. Food related events as the leading cause constituted 40%, seconded by drug in 24% and third by insect bites in 7%. Causes in 6 cases (17%) could not be determined. Shellfish and seafood accounted for most of food related events. Peanut and tree nut accounted for 3 cases. The most severe reactions with significant cardiovascular compromise were found to be related to food in 3 cases (fish 2, shellfish 1), and to drug in 3 cases (2 over the counter medication and one herbal medicine). In 4 of these severe cases, causes could not be determined. For a subgroup analysis, the patients were grouped according to triggers as Food allergy vs. Drug Allergy vs. Others. The demographic and clinical parameters were tabulated (Table 4b).

Use of Emergency Medications

In the community, there was limited self-care management. Four cases received antihistamine, 2 had oral antihistamine and 2 received intramuscular antihistamine at general practitioners' clinic.

At the Accident and Emergency Department, 10% of patients received adrenaline (intramuscular route IM 4, subcutaneous SC 4 and intravenous IV 2). Antihistamine was given to 68% of cases (IV 37, oral 21, IM 10). Systemic steroid was given to 38% of cases (IV 30, oral 8). Inhalation bronchodilator was given to 13% of cases. Fluid resuscitation was given in 8 and oxygen supplementation was required in 3 cases.

After transferal to paediatric wards, further treatment consisted of adrenaline in 10 cases (IM 5, SC 4, IV 1); antihistamine in 97 cases (oral 65, IV 32), and systemic steroid in 30 cases (IV 22, oral 8). Further fluids were given in 9 cases. Bronchodilator was required in 6. Oxygen supplementation was continued in 3 cases.

In summary, a third of the cases were treated with intravenous systemic steroid as the preferred route. Adrenaline was given in 10 cases with significant cardiovascular compromise. Intramuscular and subcutaneous routes were used in similar proportion in this series.

Allergy Assessment

The aetiology was defined by clinical evaluation alone in 56 cases. Immuno-Cap RAST were offered in 20 cases. Skin prick test were offered to 7 cases. Provocation test was applied in 2 cases. Others tests (total IgE and C1-inhibitor) were applied in 16 cases. Some cases had more than one type of evaluations. All except one patient had known allergen prior to the presenting episode.

Long Term Care and Risk Reduction for Severe Systemic Allergic Reaction/Anaphylaxis

Majority (75%) of patients were given verbal instruction for risk reduction of future severe systemic allergic reaction/ anaphylaxis. The exact content of such verbal advice was not assessed. Written instruction or action plan was offered to 6 cases and all belonged to a single institution. Eleven patients were offered adrenaline auto injector but 2 parents declined the offer. Subsequent general clinic follow up was arranged for 28% of cases. Eleven cases were referred specifically for follow up by allergy specialists if this was available in the same hospital or to other hospitals where appropriate.

EpiPen^® Availability at Hospital Pharmacy

There was a varied practice among different units. Seven out of 12 units were not providing this at their pharmacy but could prescribe for purchase at community pharmacy when indicated. Four have limited availability and provided it as patient purchased item at a cost of 600-900 dollars/ item. Only one unit has been providing it as a regular item. Two families declined the medical recommendation of carrying an EpiPen^® from units where they have to purchase it at cost.

Discussion

This first ever territory-wide study audit on management of childhood severe allergic reactions provided some unique information on the current management status in public hospitals but at the same time raised several pertinent areas of concern and pressed for improvement in the management of this life threatening condition. Statistically, we had no single mortality over the two year period, but a significant proportion of patients had severe symptoms that merited prompt resuscitative measures. We had 5-7 folds less admissions for these severe allergic reactions as compared to UK. We had not observed an increasing trend of admissions as reported in Western nations (e.g. UK and Australia).^8-10 Nonetheless, under-reporting and miss-classification could not be completely excluded. We should remain vigilant as we are witnessing an increasing globalisation and westernisation. In general, Hong Kong is a city of high atopy prevalence¹¹ and local questionnaire survey⁴ confirmed that we have a comparable rate of food allergy.

To Improve Risk Management

The impression was that we tended not to commit to a diagnosis of anaphylaxis unless overt cardiovascular symptoms were detected. Such preposition with significant implication in attitude and practice of anaphylaxis care should best be assessed by a prospective study.

Ideally, every case admitted with a diagnosis of anaphylaxis or significant angioedema should be thoroughly investigated for the possible allergen and assessed the risk of recurrence. Patient with respiratory and cardiovascular symptoms should best be equipped with auto-injector adrenaline together with a written instruction of action plan (Appendix II, this is an example format which one of participated hospital is currently using. It is not endorsed by other hospitals or COC paediatrics at the writing of this paper).

To Improve Clinical Care

The current management seemed effective in the prevention of severe complications and mortality. We were unable to assess the efficiency (how quick) of our management in the reversion of the anaphylactic episodes. The current international consensus is that adrenaline is the medication of choice for anaphylactic episodes.^9-11 Intramuscular adrenaline route is preferred.^12-15 Prompt injection of adrenaline has been associated with better outcome.^13,14 Other medications such as antihistamines, inhaled asthma medications, or steroids that are commonly used by physicians in treating anaphylaxis should not be regarded as first-line medications.^9-11 Our study showed that adrenaline auto-injectors were infrequently prescribed and used in our practice.

There was a significant proportion of undetermined cause for this cohort and yet they did not receive full assessment by a paediatric allergist/immunologist. A more detailed assessment may enable better clarification of allergens and implement probable avoidance measures. It is also of prime importance to empower the patient with appropriate knowledge and skill in dealing with severe allergic reaction is important in reducing long term morbidity and even mortality.¹⁵

Affordability and Availability of EpiPen^®

EpiPen^® is not a registered pharmaceutical product in Hong Kong. It is currently prescribed under named-patient basis and import licenses have to be applied via registration with Pharmaceutical Service of Department of Health (http://www.psdh.gov.hk/eps/index). According to a solicited data from Department of Health (Table 5), there has been a 2 fold increase in such prescriptions over a 3-year period. This included adults and children from both public and private sectors. This study raises concerns about lack of availability and affordability of adrenaline auto-injectors in Hong Kong public health care sector. Adrenaline auto-injector is not a drug item enlisted under 'safety net' for the poor. Some of our patients encountered difficulties in insurance reimbursement for purchasing EpiPen^®. At times, syringe and ampoule of adrenaline was provided instead of auto-injectors which may pose difficulties for untrained people to administer correctly in emergency situation.

Limitation

This was a multi-centre retrospective analysis and thus pertinent clinical information was not documented completely in records at times. It would also be difficult to retrieve accurate data for assessing the timing of biphasic reaction by retrospective chart review. It was arbitrarily defined as after initial stabilisation at emergency room, but required a second dose of adrenaline in ward. We acknowledge it is not the ideal way to define biphasic reaction. There were problems of misclassification, under-diagnosis and under-reporting of systemic allergic reactions with our existing practice. It is understandable as we are lacking local guidelines or bench mark standards to refer to on this condition. The understanding of exact etiology was hampered by the lack of comprehensive allergy evaluation. As the primary objective of this study is to review the practice of the Department of Paediatrics, but we do reckon emergency room's practice forms an integral part of anaphylaxis management. However, this review is limited by not including information from the emergency departments. More importantly, the search strategy that we adopted must have missed patients who attended Accident and Emergency Department of Hospital Authority hospitals but not subsequently hospitalised. In addition, the case definitions for anaphylaxis are much more variable than what was revealed by the limited codes enlisted in the current study. Our strategy might miss a fair number of patients with anaphylaxis with respiratory (upper or lower) involvement and urticaria but no cardiovascular compromise. We should try to include the ICD-9 codes for diagnoses such as 'asthma', 'food allergy', 'drug allergy' and 'urticaria' as additional search criteria in future study. To circumvent the above limitation, a proper prospective designed clinical audit in collaboration with COC in A&E will yield more accurate case identification and data retrieval.

Conclusions and Recommendations

This retrospective chart review identified a few potential shortcomings of current practice that warranted further prospective study. There were perhaps discrepancies of operational definitions and diagnostic criteria of anaphylaxis among different units. Paediatrician's preference for steroids and antihistamines, despite the current lack of evidence of their effectiveness might have resulted in possible under-use of adrenaline. There has been a lack of proper evaluation and after care of patients presenting with severe allergic reactions in current HA paediatric service.

It is highly advisable to have a local practice guideline or to endorse an international guideline with due consideration of local constraints and clinical practice. It should preferably include an operational definition of anaphylaxis for proper diagnosis and standard management pathway. It should also address the indication and prescription of auto-injectable adrenaline (e.g. EpiPen^®), which is an important resuscitative medication for treating severe allergic reactions in the community. The after care and allergy assessment of first documented anaphylaxis should also be discussed. It is vital to implement measures of improving long term risk reduction in anaphylaxis such as providing comprehensive allergy evaluation, appropriate preventive medication and written instructions. Due consultation of Emergency Departments should be made in the drafting and implementation process. We recommend to improve the availability of out-of-hospital use of auto-injectable adrenaline e.g. EpiPen^® in the Hospital Authority service. EpiPen^® should best be available in all hospital pharmacies. A registry of prescription of EpiPen^® should preferably be established. It can serve as a vital parameter for assessment of severe allergy disease burden and for subsequent audit on proper use of EpiPen^®.

Acknowledgement

We thank Professor Yu-lung Lau and Dr Betty Young for their advice. We thank Mr. Wilfred Wong for his statistical support in subgroup comparisons.

Appendix I

Appendix II

References

1. Muraro A, Roberts G, Simons FE. New visions for anaphylaxis: an iPAC summary and future trends. Pediatr Allergy Immunol 2008;19(Suppl 19):40-50.

2. Sheikh A, Hippisley-Cox J, Newton J, Fenty J. Trends in national incidence, lifetime prevalence and adrenaline prescription for anaphylaxis in England. JR Soc Med 2008;101:139-43.

3. Ross MP, Ferguson M, Street D, Klontz K, Schroeder T, Luccioli S. Analysis of food-allergic and anaphylactic events in the National Electronic Injury Surveillance System. J Allergy Clin Immunol 2008;121:166-71.

4. Leung TF, Yung E, Wong YS, Lam CW, Wong GW. Parent-reported adverse food reactions in Hong Kong Chinese pre-schoolers: epidemiology, clinical spectrum and risk factors. Pediatr Allergy Immunol 2009;20:339-46.

5. Ho MHK, Lee TL, Chow WC, Lau YL. Adrenaline Auto-injector (EpiPen^®) Use in Hong Kong: Experience of Queen Mary Hospital. HK J Paediatr (New Series) 2006;11:223-8.

6. Krugman SD, Chiaramonte DR, Matsui EC. Diagnosis and management of food-induced anaphylaxis: a national survey of paediatricians. Pediatrics 2006;118:5540-60.

7. Sampson HA, Munoz-Furlong A, Campell RL et al. Second symposium on the definition management of anaphylaxis: Summary report-second National Institute of Allergy and Infectious Disease/ Food Allergy and Anaphylaxis Network Symposium. J Allergy Clin Immunol 2006:117;391-7.

8. Muraro A, Robert G, Clark A, et al. EAACI task force on anaphylaxis in childhood: position paper of the European academy of allergology and clinical immunology. Allergy 2007;62: 857-71.

9. Baumgart K, Brown S, Gold M, et al. Australasian Society of Clinical Immunology and Allergy Anaphylaxis Working Party. ASCIA guidelines for prevention of food anaphylactic reactions in schools, preschools, and child care centres. J Paediatric Child Health 2004;40:669-71.

10. AAAAI Board of Directors. Anaphylaxis in schools and other child care settings. J Allergy Clin Immunol 1998;102:173-6.

11. Asher MI, Montefort S, Biorksten B, et al. ISAAC Phase Three Study Group. Worldwide time trends in the prevalence of symptoms of asthma, allergic rhino conjunctivitis, and eczema in childhood. ISAAC Phases One and Three repeat multicountry cross-sectional surveys. Lancet 2006;368:733-43.

12. Simon FE. First-aid treatment of anaphylaxis to food: focus on epinephrine. J Allergy Clin Immunol 2004;113:837-44.

13. Pumphrey RS. Lessons for management of anaphylaxis from a study of fatal reactions. Clin Exp Allergy 2000;30:1144-50.

14. Sheikh A, Shehata YA, Brown SG, Simons FE. Adrenaline for the treatment of anaphylaxis: Cochrane systematic review. Allergy 2009;64:204-12.

15. Ewan PW, Clark AT. Efficacy of a management plan based on severity assessment in longitudinal and case-controlled studies of 747 children with nut allergy: proposal for good practice. Clin Exp Allergy 2005; 35:751-6.