Anaphylaxis in an emergency care setting: a one year prospective study in children and adults

Anaphylaxis is a severe, life-threatening generalized hypersensitivity reaction [1]. The lifetime prevalence of anaphylaxis in the general population is estimated to be about 0.05–2.0% [2] with a variable Incidence Rate (IR) from 1.5 to 7.9 cases per 100,000 person years in Europe [3] to 42 cases per 100,000 person years in the United States [4]. In the emergency department (ED), anaphylaxis has been reported in 0.04–0.96% of all contacts [5‐7]. Overall, the case fatality is reported below 0.0001% [3].

The most severe cases of anaphylaxis are usually related to symptoms from the cardiovascular and respiratory systems [8]. Food has been reported as the most common elicitor in children, while drugs and venom are more frequent in adults [2, 3, 9]). Nevertheless, the diagnosis of anaphylaxis may be difficult in patients with many symptoms, without a certain history of allergy and in the absence of symptoms from the skin and mucosal tissue [1, 10, 11]. Moreover, the elicitors may vary according to age, sex and geographical area [1, 3]. First line treatment of anaphylaxis is adrenaline [1, 12], but previous studies have reported that only a low proportion of anaphylaxis patients receive this treatment [6, 9, 13, 14].

The acute onset of anaphylaxis, the variable symptoms and its transient and unpredictable nature may complicate prospective studies. As a consequence most of the current knowledge on anaphylaxis is derived from retrospective or register based studies [1‐4, 6‐12]. To our knowledge, only few studies have assessed anaphylaxis prospectively [5, 13, 15]. A retrospective or register based study is likely to underestimate anaphylaxis due to misclassification in diagnosis code, unclear patient history in the ED or the symptoms may have resolved spontaneously, when the patient arrives in the ED or by prehospital treatment. Thus, there is still a lack of evidence as regards the incidence, severity and epidemiological characteristics of anaphylactic reactions.

The aim of this prospective study was to evaluate patients with symptoms, diagnosis and treatment suggestive of anaphylaxis at admission to the emergency care setting. Our primary objective was to estimate the period prevalence and the Incidence Rate (IR) of anaphylaxis during 2013–2014 in children and adults admitted to the emergency care setting of Odense University Hospital. The secondary objective was to describe the severity of the anaphylactic reaction, symptoms, suspected elicitors and the pharmacological treatment administered.

We conducted a prospective, non-interventional study in all patients seen at the ED and the Acute Pediatric Ward (APW), Odense University Hospital (OUH), Denmark, during 1^st May 2013–30^th April 2014.

OUH is a 1000-bed teaching hospital representing all specialties. The population served by the ED and APW consists of four well-defined municipalities with a mixed rural-urban population of 288,587 persons (adults n = 240,070; children n = 48,517), representative for the general population. The ED attends patients ≥ 15 years of age and the APW children (0–14 years old), with a 24-h acute medical care.

In the prehospital setting, the response to an acute request of prehospital assistance is a two-tiered system, in which an ordinary ambulance manned with two emergency medical technicians (EMTs) is supplied by the Mobile Emergency Care Unit (MECU). The MECU consists of a rapid response car manned with a specialist in Anesthesiology and an EMT [16]. The ambulance can be called directly or the patients can attend either their general practitioners or a doctor on call, and then be referred to the emergency care setting in relevant cases.

Data on the population living in the hospitals catchment area during our study period were collected at the StatBank Denmark website (http://​www.​statistikbanken.​dk; accessed August 2016).

Numbers of individuals at each stage of the study and reasons for exclusion are presented in Fig. 1. Of these, 180 (female 55%, male 45%) fulfilled the criteria for the diagnosis of anaphylaxis and were included in the study; 157 (87%) adults with a median age of 45 years (IQR, 31.5–60.5) and 23 (13%) children/adolescents with a median age of 9 years (IQR, 2–15).

The proportion of individual patients with anaphylaxis in the APW was calculated to 0.4% (95% CI: 0.2–0.7) of all contacts and in the ED to 0.3% (95% CI: 0.2–0.3) of all contacts. The period prevalence of anaphylaxis in OUH’s catchments area was estimated to 0.04% (95% CI: 0.03–0.07) in children and 0.06% (95% CI: 0.05–0.07) in adults. Fifteen patients (children n = 3, adults n = 12) were seen more than once during the study period representing eighteen extra contacts in the emergency care setting.

Among the 180 patients included in the study, 110 (children n = 13, adults n = 97) experienced anaphylaxis for the first time giving an IR of 26.8 cases per 100,000 person years (95% CI: 14.3–45.8) in children and 40.4 cases per 100,000 person years (95% CI: 32.8–49.3) in adults.

One fatality due to anaphylaxis was registered during the study period. This fatal case represented 0.5% of our study population and a mortality rate of 0.3 cases per 100,000 person years in our study period. Since the patient died before arriving to the ED we chose to describe the findings independently and therefore this case is not included in further analysis.

This is one of the first prospective studies on the epidemiology of anaphylaxis in children and adults at a large university hospital, where the patients are identified not only based on diagnosis code but also on history, symptoms and treatment and thereafter classified according to international diagnosis criteria for anaphylaxis.

We found that the proportion of anaphylaxis in the emergency setting is in line with previous epidemiological studies in the ED [5, 6, 14] and the APW [7]. Moreover, the result suggests that in the group of adults, anaphylaxis presentation in the ED is as frequent as ST-elevation myocardial infarction [19].

In contrast, we found a higher IR of anaphylaxis (40.4 cases per 100,000 person years), compared to previous studies in Europe, where the IR is reported to be 1.5–7.9 cases per 100,000 person year [3], and in Denmark where the IR is reported to be increasing from 3.2 cases per 100,000 person year late in the 80’s [20] to 6.46 cases per 100,000 person year during 1995–2012 [21]. The difference could be explained by different inclusion criteria in the studies, and more specific the diagnosis codes included. In both the retrospective and register based studies of Soerensen et al [20] and Jeppesen et al [21], only the cases having diagnosis codes related to “anaphylactic shock” were included. In our study, records from 42 allergy related diagnoses (ICD-10) were reviewed, as well as all records with administered treatment of adrenaline, antihistamines or glucocorticoids to find as many cases as possible. Finally, the WAO/EAACI criteria were used for our final diagnosis of anaphylaxis. Similar differences were previously described by Bohlke [22], who could identify six times more anaphylactic episodes among children and adolescents between 1991–1997, when incorporating other allergy diagnosis besides the diagnosis codes specific for anaphylaxis. In the same way, a new retrospective and register based study of Lee et al [4] including many anaphylaxis related diagnosis (ICD-9), reported also a high IR of anaphylaxis in 42 cases per 100,000 person year, in line with our results.

Concerning the clinical characteristics of anaphylaxis, our results are in line with previous studies: the high proportion of moderate to severe anaphylaxis cases regardless of the severity score applied [9, 13],skin and respiratory symptoms as the most prevalent symptoms [5, 14, 23] and cardiovascular symptoms being more prevalent among adults compared to children [11]. It is important to highlight that 7% of the adults in our study population did not have symptoms from the skin and mucosal tissue, whereas previous studies described the absence of skin symptoms in up to 20% of anaphylaxis cases [1, 10, 11]. The latter may represent a diagnostic challenge, especially when it is not possible to collect an appropriate history of symptoms or elicitors as it may be the case with children, elderly or unconsciousness patients.

Age related differences in the elicitors profile have also been described in other studies in Europe and North America [2, 3, 7, 9]. These differences could be explained by the fact that adults are more exposed to drugs and insect stings than children/adolescents. Furthermore, food allergies are more common in children and the prevalence decreases with increasing age [24, 25].

Finally, our results about treatment of anaphylaxis substantiate those from others studies worldwide, where glucocorticoids and antihistamines seems to be administered almost as a routine, and adrenaline is administered in a lower proportion [6, 9, 13, 14, 23]. This is in contrast to international guidelines [1, 12]. Adrenaline as the first-line drug for anaphylaxis was administered in only 25% of the 173 patients with moderate to severe anaphylaxis in our study group. Whether these findings reflect the assessment of the physician on charge or insufficient treatment of the anaphylactic reaction or both cannot be clarified. However, the low degree of administration of adrenaline may be related to various factors. Firstly, the fact that anaphylaxis is mostly a prehospital condition with rapid onset, where the only possibilities to get the treatment immediately is, that the patient is equipped with an adrenaline auto injector (as was the case for 10 of our patients) or that the patient get medical help immediately (as 66 of our patients did, but only 27 were treated with adrenaline). This can probably explain why adrenaline was mostly administered at prehospital level in our study. Secondly, the improvement or even disappearance of the symptoms may occur spontaneously in some cases or because the allergen in question is removed (as may be the case in food induced anaphylaxis after vomiting). Besides, in some cases pre-existing cardiovascular conditions may have affected the decision on whether to use epinephrine or not.

Adrenaline is recommended to treat anaphylaxis due to its effect on α-1, β-1 and β-2 receptors [12] and is the only drug effective for all symptoms in anaphylaxis including hypotension and the severe respiratory symptoms [12]. The progression is fast, the course unpredictable and delayed injection of adrenaline has been described to be associated with fatal anaphylaxis [26, 27]. These facts support the importance of the prompt administration of adrenaline as well as the prescription of adrenaline auto injectors at the discharge from the emergency care settings and the reference of the patients to further allergological investigation. Standardized educational interventions including treatment algorithm and lectures on recognition, grading and management of anaphylaxis among personal at prehospital level and the staff at the ED and APW, could improve the timely administration of adrenaline.

A limitation of this study is that only patients who were referred to the Allergy Center and gave consent to participate in the study were included. Therefore, our estimates are minimum figures as we cannot rule out that some of the patients who declined participation, or were excluded, could have had anaphylaxis. Furthermore, some contacts may have been missed since some patients living in OUH’s catchments area may have attended other hospitals during the study period for anaphylaxis. Also, a potential source of error is the possibility of cases of fatal anaphylaxis at pre-hospital level not being registered in the MECU or ED. Finally, the elicitor profile is based on the patient history rather than on diagnostic work-up.

The strengths of our study are the prospective inclusion of the patients with a broad search on possible anaphylaxis cases comprising the patient history, the administered treatment and a wide list of possible IDC-10 diagnoses related to allergy and anaphylaxis. Besides this, our study is based in a population that is representative for the Danish general population. Furthermore, due to a carefully interview at the Allergy Center and use of patients files from the ED and APW, it was possible to classify all patients according to established diagnostic criteria of anaphylaxis and a severity score.

In this study focus was on the assessment and management of anaphylaxis at the admission to the emergency care setting. In the future, this patient cohort will be followed up after allergological diagnostic work-up to verify the elicitors along with evaluation of co-morbidity and co-factors in anaphylaxis.

Anaphylaxis occurred with a higher incidence compared to previous studies but with the same elicitor profile. Skin and respiratory symptoms were the most frequently observed, but up to 7% of cases in adults occurred without skin manifestations. The main elicitor of anaphylaxis in children was food, while in adults drugs and venom were the main elicitors. Adrenaline was administered significant less than glucocorticoids and antihistamines. Our findings reflect the need for standardized anaphylaxis criteria among physicians treating anaphylaxis as well as the implementation of anaphylaxis guidelines regarding diagnosis and treatment.