Should adrenaline be used in patients with hemodynamically stable anaphylaxis? Incident case control study nested within a retrospective cohort study

Although adrenaline (epinephrine) is a cornerstone of initial anaphylaxis treatment, it is not often used. We sought to assess whether use of adrenaline in hemodynamically stable patients with anaphylaxis could prevent the development of hypotension. We conducted a retrospective cohort study of 761 adult patients with anaphylaxis presenting to the emergency department (ED) of a tertiary care hospital over a 10-year period. We divided the patients into two groups according to the occurrence of hypotension and compared demographic characteristics, clinical features, treatments and outcomes. Of the 340 patients with anaphylaxis who were normotensive at first presentation, 40 patients experienced hypotension during their ED stay. The ED stay of the hypotension group was significantly longer than that of patients who did not experience hypotension (496 min vs 253 min, P = 0.000). Adrenaline use in hemodynamically stable anaphylaxis patient was independently associated with a lower risk of developing in-hospital occurrence of hypotension: OR, 0.254 [95% CI, 0.091–0.706]. Adrenaline use in hemodynamically stable anaphylaxis patients was associated with a reduced risk of developing in-hospital occurrence of hypotension. Adverse events induced by adrenaline were rare when the intramuscular route was used.


Results
During the study period, 761 patients presented to the emergency department (ED) and were given a discharge diagnosis related to anaphylaxis. Of these, we excluded 126 patients whose diagnosis was not compatible with our pre-specified population and anaphylaxis definition criteria. The reasons for exclusion were: 176 patients with hypotension as an initial presentation at hospital after symptom onset, 62 patients younger than 16 years; 57 patients who were transferred from another hospital or to another hospital. We were thus left with a total of 340 hemodynamically stable (defined as systolic blood pressure ≥ 90 mmHg) patients with anaphylaxis in the final analysis (Fig. 1). The mean age was 45.6 ± 15.3 years and 52.1% were female. During their ED stay, 40 patients (11.8%) developed hypotension. The median time from first medical contact at hospital to the occurrence of hypotension was 35.0 (interquartile range (IQR) 9.0-116.0) minutes. The demographic characteristics, comorbidities, symptoms, signs and initial vital signs of the patients who developed hypotension versus those who did not are summarized in Table 1. Comorbidities, allergy history and anaphylaxis history were not significant different between the two groups. The initial systolic and diastolic blood pressures (BP) at first medical contact of patients who developed hypotension were significantly lower than those of patients who did not develop hypotension (114.1 vs 129.3 P = 0.000, 70.3 vs 81.1 P = 0.020, respectively).
No mortality was observed in ether group. The length of ED stay in the hypotension group was significantly longer than those of patients who did not experience hypotension (496 min vs 253 min, P < 0.001). In addition, a higher admission rate was observed in those experiencing hypotension when compared to those who remained normotensive (40% vs 15.7%, P = 0.001). Treatments between the two groups did not show any significant difference except for use of adrenaline in hemodynamically stable anaphylaxis patient (Table 2). Adrenaline use in hemodynamically stable anaphylaxis patient was less frequent in the hypotensive group (P = 0.010).
A stepwise logistic regression analysis with backward elimination was performed to identify independent variables that could predict hypotension development. Adrenaline use in hemodynamically stable anaphylaxis patient was associated with a reduced risk of developing hypotension (OR, 0.254 [95% CI, 0.091-0.706]) ( Table 3).
There were two adrenaline induced adverse events. One patient was a 44 year old man, who presented with urticaria and dyspnea after taking ibupropen. He complained of chest discomfort after three minutes of receiving 0.1 mg adrenaline intravenous injection (whether dilution was performed or not was not confirmed) before development of hypotension. But it disappeared immediately after cessation of adrenaline and there were no abnormality on electrocardiogram and cardiac biomarkers. Another person was a 61 year old man with urticaria and hypotension. Adrenaline was administered after hypotension. Frequent ventricular premature beats were observed on his electrocardiogram after 0.1mg adrenaline intravenous injection (dilution was performed); he did not experience any symptoms. There was no cardiac biomarker elevation or any other adrenaline related adverse event.

Discussion
Adrenaline use was associated with reduced risk of developing hypotension in patients experiencing anaphylaxis who were normotensive on presentation to hospital.   There were only two adrenaline induced adverse events, and neither resulted in permanent harm. Both were associated with the intravenous route and were thus used against guideline recommendations which clearly state that the intramuscular route is the preferred option. Clearly, adrenaline would have to be used with care in patients with underlying cardiac disease, but it appears that it is being underused in patients with anaphylaxis, and that higher rates of use has the potential to improve outcomes in this group.
To our knowledge, no study to date has reported association between use of adrenaline and the occurrence of hypotension in anaphylaxis. To date, adrenaline use for anaphylaxis has been largely based on expert opinion and relatively weak evidence; in particular, there have been no controlled trials and given the ethical and logistical challenge to mounting these in anaphylaxis these are unlikely to be forthcoming 21,22 . We have therefore we believe undertaken as rigorous a study as is possible at the present time.
Confounding by indication needs to be considered, although drug-induced anaphylaxis was more prevalent in the hypotension group, there was no difference in the severity between the two groups at initial presentation. That said, in the absence of undertaking a randomized controlled trial we cannot be sure that adrenaline prevented the development of hypotension. A randomized, controlled trial would be the best approach for addressing this issue. Although it has been reported that prophylactic use of adrenaline can substantially reduce the risk of anaphylaxis with anti-snake venom 26 , we did not find similar results with our study.
The primary outcome was defined as the development of hypotension during ED stay because mortality with anaphylaxis is mainly due to cardiovascular and respiratory compromise 14 . Hypotension reflects a severe generalized hypersensitivity reaction and is associated with poor outcomes 27 . Additionally, in our study, the hypotension group showed a longer ED stay and higher admission rate than the normotensive group due to stabilization of anaphylaxis and observation for more fatal reactions. Furthermore, the definition of respiratory compromise is difficult because there is no definite and objective parameter that reflects respiratory compromise in anaphylaxis. In contrast, cardiovascular compromise can be easily assessed by measurement of blood pressure and is relatively objective.
The main limitation of our present study was its retrospective study design. Accordingly, important information concerning clinical symptoms and past history of allergies or anaphylaxis and other factors may have been omitted. We could only trace the course and outcomes of patients in the ED, so there is a chance that the development of hypotension or significant biphasic reaction was missed. However, recent studies reported that clinically significant biphasic reactions in anaphylaxis are quite rare, so it is unlikely that mortality or a significant biphasic reaction including hypotension was underestimated 28,29 . Our study showed a small difference in triggers and clinical features compared with preexisting studies, so it might be difficult to generalize our results to all kinds of anaphylaxis situations. Further prospective studies are warranted.
Not all anaphylaxis symptoms and signs occur simultaneously. In some cases, anaphylaxis symptoms begin as a minor form, progressing so rapidly that no treatment can be given before respiratory or cardiac arrest 14 . Sampson et al. 23 reported on fatal and near-fatal anaphylactic reactions to food in children and adolescents. Most of the patients presented skin and abdominal symptoms after 1 to 30 min that became severe between 20 min and 2.5 h later. The most important difference between fatal and nonfatal patients was that no patient in the fatal group received adrenaline before severe symptoms developed, whereas all patients with nonfatal reactions received adrenaline before or within 5 min of the development of severe symptoms. This clinical manifestation of anaphylaxis can be found in clinical practice. Thus, early administration of adrenaline should be considered, even if the first presentation is a mild form of anaphylaxis.
Adrenaline decreases mucosal edema, relieves upper airway obstruction, and increases blood pressure via an alpha-1 adrenergic vasoconstrictor effect. Beta-1 and -2 adrenergic receptor-mediated effects of adrenaline lead to inotropic and chronotropic effects and bronchodilation, which have beneficial effects in patients in anaphylaxis 20,21 . However, adrenaline also has adverse effects, from mild (i.e., pallor, tremor, anxiety, palpitations, headache, and dizziness) to severe (i.e., pulmonary edema, cardiomyopathy, left ventricular dysfunction, hypertension, cardiac arrhythmia, and myocardial infarction) 20,30 . These serious adverse events can occur by any route, but are most frequent after rapid intravenous infusion, an erroneous dose, or incorrect diluted adrenaline intravenous injection 8,30 . In our study, 2 of the 136 (all of adrenaline use before and after occurrence of hypotension) patients (1.5%) who received adrenaline experienced an adverse event, though neither resulted in long-lasting harm. Kanwar et al. 30 reported a 2.4% incidence of potentially life-threatening complications from adrenaline. These complications were mainly from an inappropriate dose or route, but our two cases were not due to an incorrect dose, and the adverse events were not severe. Intramuscular, intravenous, and other routes comprised 82%, 13%, and 5% of all adrenaline administration in our study, respectively. All major guidelines indicate intramuscular adrenaline as a first line treatment of anaphylaxis because of safety problem compared to intravenous route. Despite 2 cases of adrenaline induced adverse event in our study were not life threatening, intramuscular route seems safer than intravenous route. As guidelines suggest, intravenous route should be saved for those who require repeated dose of intramuscular adrenaline or show refractory shock despite first line treatment.
In conclusion, adrenaline use in hemodynamically stable anaphylaxis patients was associated with reduced risk of in-hospital occurrence of hypotension. Intramuscular adrenaline should be used as a first line treatment of anaphylaxis because intravenous route can trigger potentially serious adverse events.   1,31 . Two emergency physicians performed the review of medical charts together and reviewed the EAACI anaphylaxis guideline ahead of chart review. There were no uncertain cases for discussion. Patients were excluded if they were younger than 16 years, did not satisfy the above definition of anaphylaxis or were transferred from or to other hospitals. We also excluded patients who already manifested hypotension on first presentation to the attending physician and/or nurse. We only used initial blood pressure checked in hospital and did not considered blood pressure checked in emergency medical service because such data were not stored in EMR database and sometime were not available even at time of ED arrival. The primary outcome was the development of hypotension, which was defined as systolic blood pressure < 90 mmHg for more than 15 minutes without signs of other causes of shock except anaphylaxis progression during ED stay. Previous studies also used this primary outcome as severe form of anaphylaxis 32,33 . Our secondary outcome of interest was in-hospital mortality.
The clinical and demographic characteristics of all patients, including their age, sex, comorbidities, initial vital signs, laboratory findings, in-hospital course, treatment, and clinical outcomes, were retrieved from the CDW. The decision to perform treatment choice such as adrenaline, other adjunctive therapy was at the discretion of the treating physicians.
Statistical Analysis. The data in this study were presented as the mean ± standard deviation or median with the interquartile range (IQR) for continuous variables and as absolute or relative frequencies for categorical variables. Patients who developed hypotension during ED stay were compared with those who did not. Student t tests or Mann-Whitney U tests were used to compare continuous variables, and chi-square tests were used for categorical variables. The results of logistic regression analysis of adrenaline use in hemodynamically stable anaphylaxis patient adjusted with age, sex and other significant factors in univariate analysis are presented as odds ratios (ORs) and 95% confidence intervals (CIs). A p value ≤ 0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS for Windows version 18.0 (SPSS Inc. Chicago, IL, USA).