Discussion

Anaphylaxis is an acute, potentially life-threatening systemic allergic reaction. Measuring and evaluating epidemiological data related to anaphylaxis is an important way to identify disease burden trends and risk factors. Currently, epidemiological data resources for anaphylaxis included the purchase of epinephrine auto-injectors, national databases, primary care databases, surveys with representative samples of the general population, and hospitalization or emergency department admissions, [2; 16; 17], and hospital admissions datasets were deemed as the largest and most robust data available to understand trends in anaphylaxis[18]. Studies based on hospitalization usually used structured data such as the International Classification of Diseases (ICD)-9 and ICD-10 to identify anaphylaxis patients[19]. However, A weakness of such studies is misdiagnosis, and misclassification[16], which leads underestimate the incidence of anaphylaxis. For example, Klein and Yocum[20]conducted a retrospective analysis of patients records in the emergency department, and they identified 17 cases of anaphylaxis, only four of the 17 patients were diagnosed as anaphylaxis which could be identified by ICD-9.
In this study, we developed anaphylaxis trigger entries that contained both structured data (e.g., medical orders and diagnostic data) and unstructured data (e.g., progress notes). The use of unstructured data resulted in a 2-fold increase in the detection rate of anaphylaxis compared to diagnosis-based structured data, which definitely improved the performance of the programs. In addition, compared with the spontaneous reporting system of our hospital during the same period, 83.72% of cases of anaphylaxis were under-reported, suggesting that the active surveillance programs can significantly make up for the deficiency of the spontaneous reporting system, which was the primary adverse reaction monitoring method in China. Panesar[21] et al. showed that the incidence rates for anaphylaxis in Europe ranged from 1.5 to 7.9 per 100,000 person-years, In our study, the incidence of anaphylaxis in the Chinese population was 8.29 episodes per 100,000 person-years, which was higher than other studies[22; 23] based on electronic medical records, indicating that the active surveillance programs performed well. However, the sensitivity of our programs was still low; we analyzed 4874 medical records of discharged patients in our hospital from December 1st to 31st, 2020, recorded all suspected ADRs (according to the progress notes and diagnoses), 3 cases were identified as anaphylaxis, and only 1 case could be which definitely improved the performance of the programs monitored by the active surveillance programs. Analysis of undetected anaphylaxis in the above-discharged patients and the spontaneous reporting system revealed that 1 case did not receive drug treatment after anaphylaxis, 4 cases were treated with dexamethasone only, and 2 cases had no progress notes reflecting ”allergy”, this is why they could not be monitored by the active surveillance programs. Therefore, the management of anaphylaxis and medical record writing should be standardized to improve the sensitivity of the method.
Regarding demographic characteristics, our study showed that the incidence of anaphylaxis in females was significantly higher than in males. Combined with the gender composition of patients during the same period, the ratio of incidence of anaphylaxis in males to females was 1:2.1. Similar results of gender difference were reported by Banerji et al.[24] who analyzed 716 patients with anaphylaxis, and 71% were female. Studies showed that anaphylaxis in females was lower than in males before puberty but increased rapidly and exceeded with age, but the specific mechanism is unknown[25; 26].
Drug induced anaphylaxis has become more frequent with age, and death rates from drug induced anaphylaxis have risen 300% over the last decade[17]. Drugs involved in anaphylaxis vary according to different populations, time, geographic regions, drug usage habits, genetic factors, anaphylaxis definitions, registries of cases, and study designs.[27].In our study, the drug accounted for up to 83.54% of all anaphylaxis, and the top drug classes associated with anaphylaxis were antibacterial drugs, antineoplastic drugs, and contrast media. After comparing with the number of patients treated in our hospital during the same period, the proportion of anaphylaxis caused by antineoplastic drugs was the highest (0.06%), followed by antibacterial drugs (0.02%) and contrast media (0.02%). Oxaliplatin was the most common trigger in antineoplastic drugs, and similar results were found in the Korean population[28].In fact, Oxaliplatin induced hypersensitivity reaction had raised widespread attention [29; 30; 31], and the China National Medical Products Administration issued a revision of the package insert in August 2021[32], adding a black box warning, warning that oxaliplatin might cause allergic reactions, which could lead to death in severe cases. Antibacterial drugs, especially beta-lactams, were recognized as the major causes of anaphylaxis, previous studies showed that the incidence of anaphylaxis to cephalosporin was lower than penicillin[27; 28], and amoxicillin-containing drugs were the most frequently reported cause of anaphylaxis to the FDA[33]. However, in our study, cephalosporins were the drugs most frequently involved, probably because of prescription habits in China. As routine skin tests are not recommended before the use of cephalosporins, future research should focus on exploring the prediction method of allergic reactions with higher sensitivity and specificity.
Promptly intramuscular injection of epinephrine into the mid-thigh area is the first-line management of anaphylaxis, with or without shock, in various guidelines[5; 6]. The recommended dose in adults is 0.01 mg/kg body weight, with a maximum total dose of 0.5 mg. Besides, subcutaneous injection is not recommended for emergency treatment because of its lower onset of action [10]. Glucocorticosteroids and antihistamines are commonly used in anaphylaxis, However, guidelines recommended them as the second-line medications for anaphylaxis, and there is increasing evidence that their routine use is controversial. Glucocorticosteroids may be of no benefit or even harmful in the acute management of anaphylaxis[5]. In this study, we found that the proportion of glucocorticoids and antihistamines in the treatment of anaphylaxis was significantly higher than that of epinephrine (83.54% vs. 44.30%, 58.23% vs. 44.30%, P < 0.01), and the proportion of epinephrine used in non-anaphylactic shock was significantly lower than that of shock. In addition, the usage and dosage of epinephrine were irrational, such as the considerable proportion of subcutaneous injections of epinephrine and the huge dosage varies. Jiang et al.[11] also showed the significant underuse, as well as the inappropriate usage and dosage of epinephrine and the unreasonable high use of glucocorticoid in China. Therefore, it is urgent to improve the management and treatment of anaphylaxis by medical staff to reduce the death caused by anaphylaxis.
Our study has some limitations. First, this was a single-center study,the formulation of medical orders triggers was based on the prescribing habits of doctors in our hospital, which had low external validity. When applied to other hospitals, some items of the trigger need to be modified. In addition, there was a certain missed detection rate in our active surveillance programs due to the low sensitivity, and this might lead to the missed detection of anaphylaxis, which had a certain impact on the results.