Disposition of emergency department patients presenting with angiotensin-converting enzyme inhibitor-induced angioedema

Abstract

Background

Angiotensin-converting enzyme inhibitors (ACEI) are the most common cause of drug-induced angioedema in the United States. Our primary objective was to provide descriptive evidence regarding emergency department (ED) disposition of ACEI-induced angioedema patients. Our secondary objective was to evaluate unique patterns in those with ACEI-induced angioedema at a tertiary referral center, including demographics, details of those requiring intubation, length of inpatient stay, and allergy documentation.

Methods

This was a retrospective study evaluating all cases of ACEI-induced angioedema at a large, regional academic medical center. We performed a medical record review to identify patients with ACEI-induced angioedema who presented to the ED from January 1, 2016, to December 31, 2022. A structured data abstraction process was utilized to select patients of interest, followed by descriptive statistics, chi-square tests and odds ratios for categorical data, and Kruskal Wallis tests for continuous data.

Results

A total of 637 unique patient encounters met potential inclusion. After a substantial, standardized review, 94 patients met inclusion. During the study period, there were 94 patients presenting to the ED who were diagnosed as having angioedema secondary to an ACEI (90 patients) or angiotensin receptor blocker (ARB) (4 patients). Overall, 53 patients (56.38%) improved during their ED stay, and of those, 32 patients (60.38%) were discharged home. None of the 12 patients that worsened were sent home from the ED. Those who were discharged from the ED with a median stay of 4 h had no increased risk of return to the ED versus those who were admitted. The only treatment found to have statistical association with disposition was intramuscular epinephrine. Only 13 of 43 ED discharged patients (23.64%) had their ACEI/ARB documented in their allergy listings compared to 42 of 51 patients (76.36%) of admitted patients, odds ratio of failure to document was 0.0929 (95% CI, 0.0352 to 0.24512). Only one patient out of the total 94 returned to the ED due to recurrent angioedema symptoms during our study period and was subsequently discharged.

Conclusion

Our results indicated that at a large, tertiary care referral center, patients with ACEI-angioedema who did not exhibit severe symptoms were safely discharged home with low risk of early return visit to the ED. Severe symptoms were found to be any throat symptoms, voice change, drooling, worsening swelling, or complaints of dyspnea. Further, among those admitted, patients in our study had a short inpatient stay and were safely discharged with no documented early return visit to the ED.

Introduction

Angiotensin-converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARBs) are the most common cause of drug-induced angioedema in the United States (US). They account for 20–40% of all emergency department (ED) visits for angioedema annually [1]. ACEI are some of the most frequently prescribed medications, with lisinopril being named the most common antihypertensive agent in the US in 2020 [2]. Therefore, with such a frequent incidence of this pathology, life-threatening presentations of ACEI-induced angioedema can be expected [3, 4].

ACEI-induced angioedema is marked by high bradykinin levels [5]. Unfortunately, there is no proven therapy that will quickly reverse symptoms. Therapies utilized for histamine mediated angioedema are often utilized for this bradykinin mediated angioedema. These therapies, antihistamine agents, corticosteroids, and epinephrine, have not been evaluated in any prospective studies. Bradykinin pathway inhibitors like ecallantide and icatibant are expensive and have had mixed results [6, 7]. Both fresh frozen plasma (FFP) and tranexamic acid (TXA) are well-described in case reports, with the latter featured in two retrospective studies [8, 9]. One prior systematic review documented the use of ecallantide, icatibant, C1INH, FFP, TXA, and omalizumab in the relief of symptoms in patients with ACEI-induced angioedema and idiopathic angioedema. However, 29 of the 61 articles used in this systematic review were case reports [10].

Consequently, intubating patients with impending respiratory failure is the key life-saving intervention [11]. Determining who is at risk for severe illness versus who may be discharged with outpatient follow up is not well-defined. In fact, there are currently no standardized guidelines, with much of the current management recommendations deriving from expert opinion, subspecialty groups, and case reports [12,13,14,15]. Many physicians are cautious to discharge patients with ACEI-induced angioedema home, out of concern for worsening upper airway symptoms and respiratory arrest. From our literature review, as of writing there has been no study that has examined the disposition of patients with ACEI-induced angioedema from the ED.

Boarding and overcrowding in the ED continues to limit timely patient care [16]. While the approach to reducing ED overcrowding is multifactorial, emergency physicians should seek methods to reliably discharge certain patient populations that can safely follow up outpatient.

One retrospective study in 2020 evaluated ACEI-induced angioedema cases in a single, large, urban, tertiary care ED [17]. However, this study did not evaluate the number of return visits to the ED within a 72-hour period, nor did it provide details regarding ED observation versus inpatient admission. Our primary objective was to provide descriptive evidence regarding ED disposition of ACEI-induced angioedema patients, centering on a patient cohort that can be safely discharged from the ED. Our secondary objective was to evaluate unique patterns in those with ACEI-induced angioedema at a tertiary referral center, including demographics, details of those requiring intubation, length of inpatient stay, and allergy documentation.

Methods

This was a retrospective study evaluating all cases of ACEI-induced angioedema at a large, regional academic medical center and tertiary care referral center. The facility’s annual ED volume averages 79,000 patients per year. The study was evaluated by the Institutional Review Board and was deemed exempt.

We performed a medical record review to identify patients with ACEI-induced angioedema who presented to the ED from January 1, 2016, to December 31, 2022. These dates were used as a convenience sample to allow for a large cohort through all seasons. Additionally, the year 2016 was chosen as a start date because this was the first year that an electronic medical record (EMR) database was introduced at the study institution, allowing for more accurate and timely medical record review. A structured data abstraction process was utilized to select patients of interest. For statistical analysis, we used descriptive statistics, odds ratios with 95% confidence intervals, the chi-square test or Fisher’s Exact test for the frequency distributions of categorical variables, and the Kruskal-Wallis test for continuous variables. We queried the EMR for patients with an ED encounter and corresponding International Classification of Disease- 10th Revision (ICD-10) code. Only patients with an ICD-10 code of “angioedema” (T46.4 × 5 A) or “ACEI-induced angioedema” (T78.3XXA) were considered. Patient records that met any of the following criteria were excluded from further analysis: patients < 18 years old, patients found to have angioedema secondary to iatrogenic cause, ACEI-induced angioedema not documented by the ED physician caring for the patient, patients with documented genetic or chronic illness that predisposes them to angioedema (e.g. hereditary angioedema), patients found to have allergic angioedema or anaphylaxis, patients with angioedema that was not related to an ACEI or ARB but another medication. Our inclusion criteria were any patient > 18 years old found to have documented ACEI-angioedema with a clear history of ACEI/ARB usage.

A single emergency physician performed the initial chart review. After completion, two independent emergency physicians reviewed the included 94 records and were not blinded to the design of the study or exclusion and inclusion criteria. STROBE guidelines were followed to ensure adequate data collection and analysis [18]. Regarding subgroup analysis, the only source of data was documentation from the ED physician, nursing, respiratory therapy, hospitalist, critical care physicians, and anesthesia. We evaluated all this data and the conclusions reached by the providers at the time of their documentation. If there was any uncertainty regarding a case, the initial reviewer discussed this with the two independent emergency for their final approval or exclusion. If there was continued uncertainty, the case was excluded.

The following demographic features were recorded during a review of the included patients: age, sex, race, name of the causative agent, duration of therapy prior to the incident, and onset of symptoms. Further, aspects of the patient’s medical care were extracted, including the following: what agents were given in the ED, intubation status, final disposition, total time of ED stay, total inpatient stay if admitted, medical floor or intensive care unit (ICU) status, patient’s allergy listing, physical exam findings, and exam changes after ED physician reassessment. We also assessed the number of return visits to the ED, which we defined as any new ED encounter within a 72-hour period. We measured number of return visits during the patient chart review on included patients. We chose 72 h as the cutoff as this is a commonly utilized timeframe in EM literature and industry metrics [19].

Results

A total of 637 unique patient encounters met potential inclusion. After review, 94 patients met inclusion. During that time, there were 94 patients presenting to the emergency department who were diagnosed as having angioedema secondary to an ACEI (90 patients) or ARB (4 patients). A table of patient demographics can be seen in Table 1. The majority of patients were male and white, with the median age being 60 years, with an interquartile range (IQR) of 22 years. Median duration of ACEI/ARB use before swelling was 3 years, with an IQR of 3 years. Only 7 patients (7.45%) had a prior history of angioedema. Symptoms included lip swelling, 54 patients (87.45%); tongue swelling, 41 patients (43.62%); facial swelling, 22 patients (23.4%); and visible throat/neck swelling, or throat related symptoms (such as difficulty speaking), 23 patients (24.47%). Median duration of angioedema symptoms prior to arrival was five hours, the IQR was 7.5 h.

Table 1 Demographics, overall, admitted group, ED discharge group

Patient disposition

Drugs clinically identified as the culprit agent are listed in Table 2 with patient dispositions. A small majority, 51 patients (54.3%) were admitted, and 43 patients (45.7%) were discharged home. From our data, we were unable to identify a single ACEI or ARB drug that had greater proclivity to angioedema than other related drugs.

Table 2 ACEI/ARB type by disposition

Association of presenting symptoms and disposition

The only symptom group associated with hospital admission was visible throat/neck swelling or throat related symptoms such as difficult speaking (see Table 3). For patients with throat symptoms or visual throat/neck swelling, 73.9% of patients were admitted versus 54.26% overall, p = 0.0294, odds ratio 3.08 (95% CI 1.09–8.73).

Table 3 Association of presenting symptoms and emergency department disposition

Treatments

Table 4 lists treatments ranked by frequency of administration, looking for association with dispositions.

Table 4 Treatment by disposition

The only treatment found to have statistical association with disposition was intramuscular (IM) epinephrine. IM epinephrine was associated with the decision to admit the patient to the hospital for ongoing care, p = 0.0145.

Effect of symptom improvement on disposition

Documented clinical improvement with treatment in the ED was associated with discharge home from the ED (Fig. 1). The median time in the ED for discharged patients was 4.2 h, interquartile range of 1.9 h, with a minimum of 1.2 h, maximum of 9.2 h.

Fig. 1

Effect of symptom improvement on ED disposition

* Significant difference by Chi Square test, p = 0.0012

Number of return visits to the ED

Overall, one patient out of 94 total returned to the ED during our study period. This patient returned 12 h later due to continued upper lip swelling. The patient was given repeat doses of methylprednisone, famotidine, and diphenhydramine and was observed for three hours. The patient was discharged with no subsequent return visit documented.

Critical illness

Focusing on the 51 patients (54.3% of total) admitted to the hospital, 32 patients were admitted to the floor with a median hospital length of stay of 1.3 days, IQR 1.25 days, minimum 0.1 day, maximum of 4 days. Nineteen patients (20.2%) were admitted to the ICU with a median total hospital stay of 2 days, IQR 4 days, minimum of 0.5 days, maximum of 36 days. 10 of these 19 patients required intubation due to unrelieved respiratory distress, 6.3% of angioedema patients overall. Table 5 summarizes characteristics of those intubated. Regarding the culprit drugs taken by patients requiring intubation, one patient was taking valsartan, three patients were taking lisinopril, one patient was taking benazepril, and one patient was taking ramipril. Patients who were intubated were more likely to have a more rapid onset of symptoms within 4 h, as well as changes in voice or inability to control secretions. Failure to respond to traditional antihistamine medications and IM epinephrine were documented on all these patients. Anesthesia was called to intubate in 6 of the 10 cases, with 3 requiring immediate transfer to the operating room for intubation. 2 cases required emergent tracheostomy after multiple failed attempts at intubation.

Table 5 Characteristics of those intubated for ACEI-induced angioedema

Subsequent addition to allergy listings

We assessed the addition of the drug class and culprit drug being listed as an allergy after the date of the hospital admission or the emergency department visit for discharged patients (Table 6). Only thirteen of 43 ED discharged patients (23.64%) had this important reaction documented in their allergy listings compared to 42 of 51 patients (76.36%) of admitted patients having this documented the records, p < 0.0001, odds ratio of failure to document was 0.0929 (95% CI, 0.0352 to 0.24512).

Table 6 Documentation of adverse reaction on patient allergy listings

Discussion

Our results indicated that at a large, tertiary care referral center, patients with ACEI-angioedema who did not exhibit severe symptoms were safely discharged home with low risk of early return visit to the ED. Severe symptoms were found to be any throat symptoms, voice change, drooling, worsening swelling, or complaints of dyspnea. Only one patient of the 94 in our study population returned to the ED. Further, among those admitted, patients in our study had a short inpatient stay and were safely discharged with no documented early return visit to the ED.

A significant number of patients discharged from the ED did not have their ACEI or ARB recorded as an allergy in their EMR. Those discharged from an inpatient unit were more likely to have a documented allergy. This has broad implications for patient safety and preventing repeat exposure to an offending agent [20]. ED clinicians and pharmacists should take steps in their departments to prevent this potential source of patient harm.

From our data, we were unable to identify a single ACEI or ARB drug that had greater proclivity to angioedema than other related drugs. Even though the sample of patients taking angiotensin receptor blockers (ARBs) is less pronounced than those using angiotensin-converting enzyme (ACE) inhibitors in this study, it remains crucial to acknowledge the potential association between ARBs and the occurrence of significant angioedema-related illness.

Our results confirm that patients with ACEI-induced angioedema often develop symptoms several months after medication initiation (Table 1). The median duration of ACEI/ARB therapy was 3 years, with an IQR of 3 years. Clinicians should be reminded that most patients who present with ACEI-induced angioedema may do so several years after being on the offending agent, which stands in contrast to typical medication allergic reactions [4]. While patients taking ARB drugs are less frequently represented here than patients taking ACEI, the possibility exists for ARBs to be associated with angioedema resulting in severe illness.

Our data shed light on many variables that influence disposition of patients with ACEI-induced angioedema. The presenting symptoms, as well as the medications given, are of most importance for discussing safe disposition. Lip swelling and tongue swelling were the most reported symptoms (87.45% and 43.61%, respectively). However, the only symptoms associated with hospital admission were any throat or neck-related symptoms (e.g. pharyngeal swelling, drooling, uvular swelling, neck swelling, or difficulty swallowing or speaking).

Most of our patients received medications that are traditionally administered for treatment of histaminergic angioedema (i.e. methylprednisolone, diphenhydramine, famotidine). The role of these medications is unclear and previous studies have demonstrated they are associated with minimal response [4]. Additionally, other agents including icatibant, TXA, and fresh frozen plasma were given. None of these agents were associated with a significant difference in discharge or admission. Notably, the administration of IM epinephrine was the only treatment associated with a statistically significant increased rate of admission. All patients who received IM epinephrine were noted by the ED physician to have worsening symptoms upon reassessment, prompting admission. Likely the use of IM epinephrine is a marker for severity of presenting illness, rather than a cause for patient worsening.

There is an abundance of case reports, case series, and small retrospective studies on various treatment options utilized for ACEI-induced angioedema, but none of these provide the clinician with valuable information on whether a patient can be safely discharged. A retrospective study in 2019 found similar characteristics of ED patients with ACEI-induced angioedema, however the number of return visits to the ED and characteristics of admitted and discharged patients was not evaluated [17].

We found that those with documented clinical improvement upon reassessment by the ED physician were associated with higher rates of discharge home from the ED. An optimal time frame of observation has not been established for those with ACEI-induced angioedema. Of those admitted and not intubated, the median duration length of stay was 1.3 days, with a minimum of 0.1 day and maximum of 4 days. When compared to inpatient admission, a median time in the ED for discharged patients of approximately 4 h, with only one patient in the ED group returning within a 72-hour period.

In the current age of ED overcrowding, the ability to safely discharge patients is of critical importance. Further, given the lack of inpatient bed availability, reducing admissions improves ED throughput as the opportunity cost of boarding can be high [21]. Based on our data and considering both patient and operational factors in the ED, our results suggest that discharge from ED is appropriate in certain patients with a low risk of deterioration. In particular, patients observed for at least 4 h, lacking any “severe symptoms”, defined as difficulty breathing, speaking, drooling, or throat symptoms, can be safely discharged with a low risk of return to the ED.

Limitations

Our study has several limitations, most notably being a retrospective design. There is always the possibility that some ACEI-induced angioedema cases were missed during data collection. Inaccurate coding is a realistic concern that would make it difficult to locate cases using ICD-10 codes. Additionally, study chart investigators were not blinded to the objective of this study. Given the retrospective design, patient symptoms and exam findings may not be documented in a standardized fashion. For example, we did not find any documentation of abdominal or peripheral extremity swelling in our study population. Dyspnea was inconsistently documented and therefore unreliable.

We cannot completely be sure patients did not have an early return visit to an ED at another institution. Our EMR is not connected to other local, out-of-network hospitals, and therefore the possibility exists that a discharged patient could have returned to the ED within 72 h at a different hospital. However, this is unlikely given that this is the only tertiary care referral center and academic hospital in the region, with the next nearest referral center approximately 180 miles away.

Another limitation is our sample size. External validity is limited given that this study was performed at a single center in a particular region of the US. However, the ED annual volume is similar to other tertiary care centers with teaching designations in the US [22].

Conclusion

Our data show that carefully selected ACEI-induced angioedema patients presenting to the ED were safely discharged, with a low number of return to the ED within 72 h. No particular symptom or sign predicted the need for intubation, but those who were discharged from the ED with a median stay of 4 h had no increased risk of return to the ED versus those who were admitted. A significant number of patients discharged from the ED did not have their ACEI or ARB recorded as an allergy in their EMR. ED clinicians and pharmacists should take steps in their departments to prevent this potential source of patient harm. Future research should be aimed at improving drug allergy recording in the ED and its effect on patient outcomes. Additionally, studies designed to find an ideal observation time for those with ACEI-induced angioedema in the ED are warranted.