Abstract
BACKGROUND:
There is a gap in knowledge regarding the contemporary epidemiology of penicillin-associated anaphylaxis.
OBJECTIVE:
To assess whether penicillin antibiotics (PAs) had higher reporting odds of anaphylaxis compared with other medications, patterns of anaphylaxis reports attributed to specific PAs, and the frequency of reported mortality associated with anaphylaxis using pharmacovigilance methods using the Food and Drug Administration Adverse Event Reporting System database.
METHODS:
We included patients in the Food and Drug Administration Adverse Event Reporting System with reported anaphylaxis to PAs from 2013 to 2023 in this cross-sectional study. Reporting odds ratios and proportional reporting ratios were used to determine whether specific PAs had higher reporting odds of anaphylaxis.
RESULTS:
Among a total of 20,815,425 adverse events and 48,637 reports of anaphylaxis, there were 3,176 reports of anaphylaxis to PAs. Compared with other medications, all PAs had higher reporting odds of anaphylaxis. Compared with all other PAs, there were disproportionality signals for reported anaphylaxis to amoxicillin-containing medications (reporting odds ratio, 1.97; 95% CI, 1.82–2.14) and “unspecified” PAs (reporting odds ratio, 1.77; 95% CI, 1.44–2.19). We observed a higher proportion of anaphylaxis reports to all PAs (4.5%) from reports outside the United States (US) compared with reports from the US (2.3%, P < .0001). US reports demonstrated higher reporting odds for anaphylaxis to nonaminopenicillins and piperacillin-tazobactam, whereas non-US reports demonstrated higher reporting odds for amoxicillin-containing medications. There were 139 reports of anaphylaxis to PAs associated with death.
CONCLUSIONS:
PAs had higher reporting odds of anaphylaxis compared with all other medications in the Food and Drug Administration Adverse Event Reporting System database. We observed regional differences in reporting odds to specific PAs when comparing US and non-US reports. Fatalities reported following anaphylaxis were rare.
Keywords: Penicillin allergy, Beta-lactam allergy, Adverse drug reactions, Drug allergy, Anaphylaxis, Pharmacovigilance
INTRODUCTION
Penicillin antibiotics (PAs) are critically important to public health given their effectiveness in treating many common and community-acquired infections, narrow spectrums of activity, and reassuring safety profiles. However, penicillin allergy is also the most common patient-reported antibiotic allergy.1 Studies have demonstrated that although more than 10% of the population carries a penicillin allergy label, more than 95% of such patients tolerate PA.2 Overall, there is a low prevalence of true penicillin allergy, such that less than 1% of any given population is truly allergic.2 When penicillin allergy labels are evaluated, studies suggest that the prevalence of penicillin allergy and the specific culprit PA verified by testing may also vary regionally.3 Few studies have investigated the epidemiology of penicillin anaphylaxis, a severe life-threatening reaction, which is thankfully rare.4–6 When there is a recent history compatible with anaphylaxis temporally associated with a PA, there is a much higher probability that this represents a true, verifiable allergy.4–11
It has been historically difficult to study penicillin-associated anaphylaxis, especially in national and multinational studies, because determination of incidence and risk is reliant on vast amounts of drug administration data, paired with data on drug reactions.4–6 To our knowledge, the largest and most recent study of penicillin-associated anaphylaxis in the United States was a single-site very large retrospective population-based study assessing the incidence of adverse drug reactions (including anaphylaxis) among patients receiving PAs.4 It used electronic medical record data collected from 2009 to 2017 at Kaiser Permanente Southern California. Although this study did not involve drug allergy testing to verify allergies, the design was greatly strengthened by allergist adjudication of severe reactions (including anaphylaxis). In this study, Liang et al4 found that overall penicillin-associated anaphylaxis was quite rare, occurring in 1 of 255,320 oral and 1 of 123,792 parenteral PA courses.
Recently, an epidemiologic study using the Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) demonstrated that cephalosporins exhibited disproportionate reporting of anaphylaxis compared with other medications in the database.12 In particular, some cephalosporins only known to be available and administered intravenously (IV) were disproportionately associated with anaphylaxis reports compared with other cephalosporins. There is a similar gap in knowledge of contemporary epidemiology of penicillin anaphylaxis, including studies addressing whether PAs are disproportionately associated with anaphylaxis (compared with other medications), whether there are specific PAs that are associated with a higher reporting of anaphylaxis, whether patterns of anaphylaxis to specific PAs may differ between the United States and other countries, and the proportion of anaphylaxis reports associated with PAs that are associated with mortality.
These research questions can potentially be addressed using large national event reporting systems, or national medical records systems. Therefore, to address this gap in knowledge, we conducted a pharmacovigilance study of anaphylaxis to PAs using FAERS, following previously published methods.7,12–15 The objectives of this study were to determine whether PAs are associated with higher reporting odds of anaphylaxis in the FAERS database (compared with all other medications), patterns of anaphylaxis reporting among individual PAs, and the frequency of mortality associated with reported anaphylaxis to PAs. A secondary aim was to determine whether there were differences in the proportion of anaphylaxis reports to specific PAs among reports generated in the United States compared with those generated outside the United States.
METHODS
In this cross-sectional study, we used previously established methods to review data on reported penicillin anaphylaxis and medication adverse events in the FAERS database between January 2013 and December 2023.7,12,15 Of note, the FAERS database contains medication-associated adverse events reported to the FDA by drug manufacturers, health care professionals, and patients. The FAERS database contains reports from US sources, as well as from countries abroad (non-US sources). All medications associated with a particular adverse event may be recorded. Although the data are cleaned for duplicate reports, given the nature of reporting, some duplicates may remain in the database. Only select patient-level data are available, and the clinical context is most often unknown. Firm causality between the reaction agent and the adverse event has not been established with drug allergy testing. In addition, FAERS data are curated and refined by the FDA over time. However, although the FAERS database has these inherent limitations, it is also the largest national adverse event reporting system and analysis of its data has been used in many studies for pharmacovigilance efforts and to better understand drug adverse reaction epidemiology.7,12–16 Institutional review board approval was not necessary for this study of publicly available data. We followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guideline.17
Anaphylaxis reports were identified in FAERS using the search terms “anaphylactic reaction,” “anaphylactic shock,” “anaphylactoid reaction,” and “anaphylactoid shock.”7,12 We included PAs available in the United States and those with at least 100 adverse events reported. We included PAs whose standard formulation uses beta lactams resistant to beta lactamase activity such as clavulanic acid or sulbactam, but we did not include combination 3/4-drug regimens for Helicobacter pylori. We also extracted available data on patient-level demographic and reporting characteristics, including age, sex, year, reporter type, and region. Reporting region was reported as “domestic,” “foreign,” or “unknown” in FAERS, which we refer to in this study as “US,” “non-US,” and “unknown,” respectively.
The primary outcomes of our study were the reporting odds ratios (RORs) of anaphylaxis to PAs compared with all other medications in the database. Secondary outcomes included the proportional reporting ratios (PRRs) of PAs compared with all other medications in the database, the RORs and PRRs for specific penicillins compared with all PAs, and the proportion of PA anaphylaxis reports that were associated with mortality. We also assessed the proportion of reported anaphylaxis to specific penicillins in US reports compared with non-US reports, and performed a stratified analysis examining RORs for specific penicillins compared with all other PAs in US and non-US reports separately. If we observed different disproportionality signals in the stratified analyses, we examined country-level data for PAs associated with higher reporting odds of anaphylaxis.
Statistical analysis
Statistical analysis was performed from November to December 2024. Descriptive analyses were conducted on available sociodemographic and patient-level data for patients with reported anaphylaxis associated with penicillins. For categorical variables, bivariate analyses were performed using χ2 tests and Fisher exact tests, as appropriate. Bonferroni correction was applied for multiple comparisons. Of note, the FAERS database includes adverse reactions to any medication or supplement reported to the FDA. This includes, for example, chemotherapeutic agents, biologics, antibiotics, opiates, and over-the-counter medications (eg, ibuprofen). Consistent with previous studies, we computed the RORs and PRRs for anaphylaxis pertaining to individual PAs in comparison to all other reported drugs in the FAERS database, and performed a separate analysis computing RORs and PRRs for anaphylaxis pertaining to individual penicillins in comparison to all PAs.7,12,15 The ROR quantifies the odds of reporting a specific adverse event connected to a specific medication relative to the reporting odds of the same adverse event for all other medications in the database. The PRR reflects the proportion of anaphylaxis reports for a specific drug divided by the proportion of anaphylaxis reports for all medications in the database (the comparator group).12,15 We considered the following significant: ROR with a lower limit of the 2-sided 95% CI greater than 1, and PRR above 2.12 Of note, it is important to recognize that the FAERS database does not have a denominator of the total number of drug administrations. Therefore, the ROR and PRR do not estimate relative risk but rather are epidemiologic measures indicating a medication’s relative and specific hazard of an adverse event compared with other medications in the database.
Lastly, during the study period we worked on developing a novel method for deduplicating FAERS reports.18 Acknowledging limitations, including that the FAERS data had undergone changes and refinement from the FDA since the initial data extraction in November 2024, we performed a second data extraction in January 2025. We performed deduplication on a subset of the database and a sensitivity analysis using these data of select disproportionality signals for quality assurance purposes. Deduplication was based on the AEOLUS pipeline, with additional manual steps for sanitization. Most statistical analyses were performed using STATA version 18.0 (StataCorp, College Station, Tex).
RESULTS
Between January 2013 and December 2023, a total of 20,815,425 adverse events were reported in FAERS, of which 48,637 (0.23%) were reported as anaphylaxis. There were 81,696 total adverse events attributed to PAs (0.39% of all adverse events in the FAERS database). Of the adverse events attributed to PAs, there were 3,176 reports of anaphylaxis (6.5% of all reports of anaphylaxis in the FAERS database). Anaphylaxis reports to PAs were reported as affecting predominantly adult (age ≥18 years) and female patients (Table I). The reports were primarily generated by health care professionals in non-US countries, with more reports generated in more recent years (Table I). Among 3,176 reports of anaphylaxis associated with PAs, the highest number of anaphylaxis reports were for amoxicillin (1,353 reports, 42.6%), amoxicillin-clavulanic acid (1,020 reports, 32.1%), piperacillin-tazobactam (323 reports, 10.2%), and penicillin G (158 reports, 5.0%) (Table II). Most of these reports were also for adult patients (age ≥ 18 years).
TABLE I.
Characteristics of patients with reported penicillin anaphylaxis in the FAERS database (January 2013 through December 2023)
| Characteristics | All penicillins (N = 3059) | Amoxicillin-containing* (N = 2336) | Ampicillin-containing† (N = 119) | Dicloxacillin (N = 4) | Nafcillin (N = 3) | Oxacillin (N = 6) | Penicillin-containing‡(N = 316) | Piperacillin-containing§ (N = 336) |
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Patient age (y) | ||||||||
| <18 | 143 (4.7) | 113 (4.8) | 7 (5.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 3 (1.0) | 20 (6.0) |
| 18–64 | 1657 (54.2) | 1351 (57.8) | 55 (46.2) | 3 (75.0) | 3 (100.0) | 3 (50.0) | 135 (42.7) | 145 (43.2) |
| ≥65 | 778 (25.4) | 543 (23.2) | 38 (31.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 67 (21.2) | 135 (40.2) |
| Unknown | 481 (15.7) | 329 (14.1) | 19 (16.0) | 1 (25.0) | 0 (0.0) | 3 (50.0) | 111 (35.1) | 36 (10.7) |
| Patient sex | ||||||||
| Male | 1,231 (40.2) | 975 (41.7) | 34 (28.6) | 3 (75.0) | 0 (0.0) | 3 (50.0) | 55 (17.4) | 178 (53.0) |
| Female | 1541 (50.4) | 1138 (48.7) | 71 (59.7) | 1 (25.0) | 3 (100.0) | 1 (16.7) | 222 (70.3) | 134 (39.9) |
| Unknown | 287 (9.4) | 223 (9.6) | 14 (11.8) | 0 (0.0) | 0 (0.0) | 2 (33.3) | 39 (12.3) | 24 (7.1) |
| Reporter type | ||||||||
| Health care professional | 2779 (90.9) | 2165 (92.7) | 115 (96.6) | 2 (50.0) | 3 (100.0) | 3 (50.0) | 241 (76.3) | 307 (91.4) |
| Consumer | 257 (8.4) | 154 (6.6) | 3 (2.5) | 2 (50.0) | 0 (0.0) | 2 (33.3) | 71 (22.5) | 28 (8.3) |
| Unknown | 23 (0.8) | 17 (0.7) | 1 (0.8) | 0 (0.0) | 0 (0.0) | 1 (16.7) | 4 (1.3) | 1 (0.3) |
| Year | ||||||||
| 2013 | 119 (3.9) | 84 (3.6) | 4 (3.4) | 1 (25.0) | 0 (0.0) | 1 (16.7) | 5 (1.6) | 24 (7.1) |
| 2014 | 154 (5.0) | 112 (4.8) | 11 (9.2) | 2 (50.0) | 0 (0.0) | 1 (16.7) | 9 (2.9) | 21 (6.3) |
| 2015 | 149 (4.9) | 111 (4.8) | 7 (5.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 19 (6.0) | 17 (5.1) |
| 2016 | 174 (5.7) | 102 (4.4) | 6 (5.0) | 0 (0.0) | 0 (0.0) | 1 (16.7) | 31 (9.8) | 37 (11.0) |
| 2017 | 133 (4.4) | 95 (4.1) | 8 (6.7) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 13 (4.1) | 20 (6.0) |
| 2018 | 393 (12.9) | 332 (14.2) | 10 (8.4) | 0 (0.0) | 0 (0.0) | 3 (50.0) | 31 (9.8) | 27 (8.0) |
| 2019 | 450 (14.7) | 369 (15.8) | 23 (19.3) | 0 (0.0) | 3 (100.0) | 0 (0.0) | 41 (13.0) | 20 (6.0) |
| 2020 | 434 (14.2) | 338 (14.5) | 8 (5.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 61 (19.3) | 29 (8.6) |
| 2021 | 316 (10.3) | 217 (9.3) | 26 (21.9) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 49 (15.5) | 35 (10.4) |
| 2022 | 346 (11.3) | 272 (11.6) | 6 (5.0) | 1 (25.0) | 0 (0.0) | 0 (0.0) | 25 (7.9) | 49 (14.6) |
| 2023 | 391 (12.8) | 304 (13.0) | 11 (9.2) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 32 (10.1) | 57 (17.0) |
| Region | ||||||||
| Domestic | 498 (16.3) | 195 (8.4) | 46 (38.7) | 0 (0.0) | 3 (100.0) | 1 (16.7) | 162 (51.3) | 102 (30.4) |
| Foreign | 2560 (83.7) | 2141 (91.7) | 73 (61.3) | 4 (100.0) | 0 (0.0) | 5 (83.3) | 153 (48.4) | 234 (69.6) |
| Unknown | 1 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 1 (0.3) | 0 (0.0) |
Values are n (%).
Data missing from the FAERS descriptive characteristics not included in ths analysis.
Amoxicillin-containing: isolated amoxicillin (N = 1353), amoxicillin-clavulanate (N = 1016).
Ampicillin-containing: isolated ampicillin (N = 79), ampicillin-sulbactam (N = 40).
Penicillin-containing: penicillin V (N = 80), penicillin G (N = 153), penicillin unspecified (N = 95).
Piperacillin-containing: isolated piperacillin (N = 15), piperacillin-tazobactam (N = 323).
TABLE II.
Total reported adverse events and reporting odds for PAs from the FAERS database (January 2013 to December 2023, N = 20,815,425 total adverse events to all medications, N = 48,637 reports of anaphylaxis to all medications)
| Penicillins | Total adverse events (N = 81,696) | Total anaphylaxis events (N = 3,176) | Total deaths (N = 283) | PRR for all drugs (95% CI) | ROR for all drugs (95% CI) | PRR for all PCNs (95% CI) | ROR for all PCNs (95% CI) |
|---|---|---|---|---|---|---|---|
|
| |||||||
| Amoxicillin combined | 49,453 | 2,373 | 99 | 21.54* (20.69–22.42) | 22.57* (21.64–23.55) | 1.93 (1.78–2.08) | 1.97* (1.82–2.14) |
| Amoxicillin | 26,287 | 1,353 | 29 | 22.63* (21.47–23.85) | 23.80* (22.52–25.16) | 1.56 (1.46–1.68) | 1.60* (1.48–1.71) |
| Amoxicillin-clavulanic acid | 23,166 | 1,020 | 71 | 19.22 (18.09–20.43) | 20.07* (18.83–21.38) | 1.20 (1.11–1.29) | 1.20* (1.12–1.30) |
| Ampicillin combined | 4,784 | 119 | 2 | 10.67* (8.93–12.74) | 10.92* (9.10–13.10) | 0.63 (0.52–0.75) | 0.62 (0.51–0.74) |
| Ampicillin | 3,316 | 79 | 0 | 10.21* (8.21–12.70) | 10.44* (8.35–3.05) | 0.60 (0.48–0.75) | 0.59 (0.47–0.74) |
| Ampicillin-sulbactam | 1,468 | 40 | 2 | 11.67* (8.60–15.84) | 11.97* (8.74–16.39) | 0.70 (0.51–0.95) | 0.69 (0.50–0.94) |
| Dicloxacillin | 310 | 4 | 0 | 5.52* (2.09–14.62) | 5.58* (2.08–14.97) | 0.33 (0.13–0.89) | 0.32 (0.12–0.86) |
| Nafcillin | 373 | 3 | 0 | 3.44* (1.11–10.63) | 3.46* (1.11–10.78) | 0.21 (0.07–0.64) | 0.20 (0.06–0.62) |
| Oxacillin | 479 | 6 | 0 | 5.36* (2.42–11.87) | 5.42* (2.42–2.12) | 0.32 (0.14–0.71) | 0.31 (0.14–0.70) |
| Penicillin combined | 8,710 | 333 | 15 | 16.47* (14.82–18.30) | 17.08* (15.31–19.07) | 0.98 (0.88–1.10) | 0.98 (0.87–1.10) |
| Penicillin V | 3,006 | 80 | 3 | 11.41* (9.19–14.16) | 11.69* (9.36–14.60) | 0.68 (0.54–0.84) | 0.67 (0.53–0.84) |
| Penicillin G | 4,267 | 158 | 12 | 15.90* (13.64–18.53) | 16.47* (14.05–19.31) | 0.95 (0.81–1.11) | 0.95 (0.81–1.12) |
| “Penicillin” (unspecified) | 1,437 | 95 | 0 | 28.35* (23.34–34.43) | 30.28* (24.59–37.29) | 1.72 (1.41–2.10) | 1.77* (1.44–2.19) |
| Piperacillin combined | 17,587 | 338 | 25 | 8.28* (7.44–9.20) | 8.42* (7.56–9.38) | 0.43 (0.39–0.49) | 0.42 (0.38–0.47) |
| Piperacillin | 895 | 15 | 2 | 7.17* (4.34–11.85) | 7.28* (4.37–12.13) | 0.43 (0.26–0.71) | 0.42 (0.25–0.70) |
| Piperacillin-tazobactam | 16,692 | 323 | 23 | 8.33* (7.47–9.28) | 8.47* (7.59–9.47) | 0.44 (0.39–0.49) | 0.43 (0.38–0.48) |
Deaths refer to deaths associated with anaphylaxis to penicillins reported in ths FAERS database.
Result considered significant (ROR with a lower limit of the 2-sided 95% CI >1, PRR >2).
PAs show higher reporting odds of anaphylaxis
We found that when compared with all other medications in the FAERS database, all PAs were associated with higher reporting odds of anaphylaxis, with the highest reporting odds for aminopenicillins, especially amoxicillin-containing drugs (Table II). Our analysis also showed higher reporting odds of anaphylaxis to amoxicillin and “unspecified” penicillins compared with all PAs (Table II). In our sensitivity analysis using a deduplicated version of select FAERS data, we did not observe major differences in the directionality of proportionality or significance, which was reassuring (Table III).
TABLE III.
Sensitivity analysis with deduplicated FAERS records to assess reporting odds of anaphylaxis to PAs in the FAERS database (January 2013 to December 2023, N = 13,986,839 total adverse events to all medications, N = 48,519 reports of anaphylaxis to all medications)
| Penicillins | PRR for all drugs (95% CI) | ROR for all drugs (95% CI) | PRR for all PCNs (95% CI) | ROR for all all PCNs (95% CI) |
|---|---|---|---|---|
|
| ||||
| Amoxicillin combined | 18.5* (17.36–19.62) | 19.67* (18.43–21.00) | 2.07* (1.82–2.37) | 2.14* (1.87–2.47) |
| Ampicillin combined | 8.32* (6.11–11.35) | 8.55* (6.21–11.77) | 0.55 (0.401–0.753) | 0.54 (0.39–0.74) |
| Dicloxacillin | 4.18* (1.06–16.56) | 4.23* (1.05–17.11) | 0.28 (0.07–1.11) | 0.27 (0.07–1.10) |
| Nafcillin | 3.51* (1.13–10.79) | 3.54* (1.13–11.05) | 0.24 (0.08–0.73) | 0.23 (0.07–0.71) |
| Oxacillin | 5.77* (2.18–15.21) | 5.87* (2.18–15.80) | 0.39 (0.15–1.03) | 0.38 (0.14–1.01) |
| Piperacillin combined | 10.44* (8.83–12.35) | 10.81* (9.08–12.87) | 0.67 (0.56–0.80) | 0.65 (0.55–0.79) |
Deaths refer to deaths associated with anaphylaxis to PAs reported in the FAERS database.
Data extraction performed separately from initial extraction.
Result considered significant (ROR with a lower limit of the 2-sided 95% CI >1, PRR >2).
Drugs implicated in anaphylaxis reports are different between US and non-US reports
Bivariate analyses showed that the frequency of reported anaphylaxis to all PAs and attributed to specific penicillins differed between US and non-US reports (Table IV). When compared with all adverse events reported to PAs, 513 (2.3%) episodes were anaphylaxis in US reports and 2653 (4.5%) episodes were anaphylaxis in non-US reports (P < .0001). Ampicillin, penicillin V, penicillin G, “unspecified” penicillins, and piperacillin-tazobactam comprised a higher proportion of US reports of penicillin anaphylaxis compared with non-US reports (Table IV). In contrast, amoxicillin-containing medications comprised a higher proportion of anaphylaxis in non-US reports (Table IV). Furthermore, when restricted to US reports, the following PAs were associated with higher reporting odds of anaphylaxis: nonaminopenicillins (eg, penicillin V and unspecified penicillins) and piperacillin-tazobactam (Table III). In comparison, when restricted to non-US reports only, amoxicillin, amoxicillin-clavulanic acid, penicillin G, and unspecified penicillins were associated with higher reporting odds of anaphylaxis (Table IV). We also observed a trend in increased reporting of amoxicillin-associated anaphylaxis among non-US reports, but not among US reports (Figure 1). We observed that among reports of anaphylaxis to amoxicillin from non-US sources, most (55%) were from France (Figure 2). Among reports of anaphylaxis to amoxicillin-clavulanic acid, most reports were from the United Kingdom (26%), France (23%), and Spain (11%) (Figure 3).
TABLE IV.
Reported cases of penicillin anaphylaxis, as well as reporting odds in the United States and abroad in the FAERS database (January 2013 to December 2023)
| Penicillins | Total cases (N = 3176) | US Cases (N = 513), n (%) | Non-US cases (N = 2653), n (%) | P value | US Cases: ROR among all PCNs (95% CI) | Non-US cases: ROR among all PCNs (95% CI) |
|---|---|---|---|---|---|---|
|
| ||||||
| Amoxicillin combined | 0.54 (0.45–0.64) | 2.67* (2.42–2.95) | ||||
| Amoxicillin | 1353 | 120 (23.4) | 1233 (46.5) | <.0001† | 0.76 (0.62–0.93) | 1.76* (1.63–1.90) |
| Amoxicillin-clavulanic acid | 1016 | 79 (15.4) | 937 (35.3) | <.0001† | 0.56 (0.44–0.71) | 1.29* (1.19–1.40) |
| Ampicillin combined | 1.24 (0.92–1.69) | 0.52 (0.41–0.66) | ||||
| Ampicillin | 79 | 38 (7.4) | 41 (1.6) | <.0001† | 1.32 (0.95–1.85) | 0.47 (0.35–0.65) |
| Ampicillin-sulbactam | 40 | 8 (1.6) | 32 (1.2) | .512 | 1.10 (0.54–2.23) | 0.61 (0.43–0.86) |
| Dicloxacillin | 4 | 0 (0) | 4 (0.2) | >.99 | 0 | 0.49 (0.18–1.33) |
| Nafcillin | 3 | 3 (0.6) | 0 (0) | .004 | 0.34 (0.11–1.06) | 0 |
| Oxacillin | 6 | 1 (0.2) | 5 (0.2) | 1.000 | 0.22 (0.03–1.56) | 0.38 (0.16–0.91) |
| Penicillin combined | 1.72* (1.42–2.07) | 0.93 (0.79–1.09) | ||||
| Penicillin V | 80 | 33 (6.4) | 47 (1.8) | <.0001† | 2.90* (2.21–3.81) | 0.33 (0.24–0.48) |
| Penicillin G | 153 | 66 (12.9) | 87 (3.3) | <.0001† | 1.03 (0.79–1.33) | 1.37* (1.10–1.70) |
| “Penicillin” (unspecified) | 95 | 63 (12.3) | 31 (1.2) | <.0001† | 1.58* (1.10–2.26) | 2.11* (1.56–2.85) |
| Piperacillin combined | 1.34* (1.07–1.66) | 0.30 (0.26–0.34) | ||||
| Piperacillin | 15 | 1 (0.2) | 14 (0.5) | .490 | 0.73 (0.10–5.29) | 0.36 (0.21–0.61) |
| Piperacillin-tazobactam | 323 | 101 (19.7) | 222 (8.37) | <.0001† | 1.35* (1.08–1.68) | 0.30 (0.26–0.35) |
PCN, Penicillin.
χ2 or Fisher exact test (as appropriate) was used to compare proportions among groups.
Result considered significant (ROR with a lower limit of the 2-sided 95% CI >1, PRR >2).
A significant P value after Bonferroni correction (P < .004).
FIGURE 1.
Trends in the number of reports of anaphylaxis associated with PAs. (A) All sources. (B) US data reports only. (C) Non-US data reports only.
FIGURE 2.
Cases of reported anaphylaxis to amoxicillin by country in the FAERS database (January 2013 to December 2023) (N = 1353). Countries included separately if 40 or more reports.
FIGURE 3.
Cases of reported anaphylaxis to amoxicillin-clavulanic acid by country in the FAERS database (January 2013 through December 2023) (N = 1020). Countries included separately if 40 or more reports.
Fatal outcomes among penicillin anaphylaxis reports
Among the 3176 reported cases of PA-associated anaphylaxis between January 2013 and December 2023, there were 142 (4.5%) anaphylaxis reports that were associated with death: 18 from US reports and 124 from other countries (P = .08) (Table V). Patterns of anaphylaxis fatality by drug between US and non-US reports were similar to overall patterns of penicillin anaphylaxis between US and non-US reports.
TABLE V.
Reported mortality associated with anaphylaxis to PAs in the FAERS database (January 2013 to December 2023)
| Penicillins | Total anaphylaxis reports (N = 3058) | Total anaphylaxis reports associated with fatality (N = 142) | US Fatal reports (N = 18), n (%) | Non-US fatal reports (N = 124), n (%) |
|---|---|---|---|---|
|
| ||||
| Amoxicillin combined | 2336 | 100 | 4 (22.2) | 96 (77.4) |
| Amoxicillin | 1353 | 29 | 3 (16.7) | 26 (21.0) |
| Amoxicillin-clavulanic acid | 1016 | 71 | 1 (5.5) | 70 (56.5) |
| Ampicillin combined | 119 | 2 | 1 (5.5) | 1 (0.8) |
| Ampicillin | 79 | 0 | 0 (0.0) | 0 (0.0) |
| Ampicillin-sulbactam | 40 | 2 | 1 (5.5) | 1 (0.8) |
| Dicloxacillin | 4 | 0 | 0 (0.0) | 0 (0.0) |
| Nafcillin | 3 | 0 | 0 (0.0) | 0 (0.0) |
| Oxacillin | 6 | 0 | 0 (0.0) | 0 (0.0) |
| Penicillin combined | 316 | 15 | 7 (38.9) | 8 (6.4) |
| Penicillin (unspecified) | 95 | 4 | 2 (11.1) | 2 (1.6) |
| Penicillin G | 153 | 8 | 5 (27.8) | 3 (2.4) |
| Penicillin V | 80 | 3 | 0 (0.0) | 3 (2.4) |
| Piperacillin combined | 335 | 25 | 6 (33.3) | 19 (15.3) |
| Piperacillin | 15 | 2 | 1 (5.5) | 1 (0.8) |
| Piperacillin-tazobactam | 323 | 23 | 5 (27.8) | 18 (14.5) |
DISCUSSION
In summary, the objective of this cross-sectional study was to use pharmacovigilance methodology to investigate the contemporary epidemiology of penicillin anaphylaxis in a large national and international adverse event reporting system (FAERS). We aimed to assess whether PAs were associated with higher reporting odds of anaphylaxis compared with other medications, whether there were specific penicillins that showed particularly high reporting odds (with any regional differences noted), and the number of cases reported to be associated with mortality. We observed that indeed all PAs were associated with higher reporting odds of anaphylaxis compared with all other medications in the FAERS database, reaffirming that this medication class may be a common cause of anaphylaxis (compared with other medications). We also found that when considering all reports of anaphylaxis associated with PAs, amoxicillin-containing medications and unspecified penicillins had the highest reporting odds of anaphylaxis. There was a remarkable difference in the pattern of reporting when comparing reports from US with reports from non-US sources. Non-US sources had a higher overall proportion of anaphylaxis reports to PAs compared with all other adverse events to PAs. The pattern of reported anaphylaxis attributed to specific penicillins was also different when comparing US and non-US reports. In particular, non—aminopenicillin- and piperacillin-tazobactam—associated anaphylaxis was reported more among the US reports, whereas anaphylaxis to amoxicillin-containing medications was reported more among non-US reports.
We believe these results have valuable implications regarding penicillin allergy epidemiology, particularly when examined alongside recent clinical studies examining the prevalence of penicillin allergy and sensitization patterns, as well as Liang et al’s aforementioned population-based study assessing the incidence of PA-associated anaphylaxis.4 Overall, the penicillin drug class appears to be a common cause of reported anaphylaxis, but there may be regional and national differences in the prevalence and patterns of IgE-mediated allergy to specific penicillins.3,19,20 In particular, our pharmacovigilance study showed that amoxicillin-containing medications were associated with higher reporting odds of anaphylaxis outside the United States, but not among US reports. This finding aligns with results from skin testing studies, which have shown predominant sensitization to aminopenicillins among patients in many European countries; this is less common in other regions of the world, including the United States.1,21–28 Likewise, our results showed that among US reports only, piperacillin-tazobactam was disproportionately associated with anaphylaxis reports. This also complements recent clinical testing studies from the United States that have suggested that selective sensitization to piperacillin-tazobactam may be clinically relevant, especially among patients with repeated exposures, such as among patients with cystic fibrosis.29
There are several possible explanations for the differential reporting patterns between US and non-US reports that we observed. Our opinion is that this differential pattern may be reflective of “locality” in drug allergy epidemiology and the available drug formulary. Although we did not directly assess this in this study, we hypothesize that differing utilization patterns and formularies in the reporting countries may lead to different sensitization and reaction patterns. Unlike our previous study examining the route of drug administration as a risk factor for reported cephalosporin anaphylaxis, we cannot draw any firm conclusions regarding route of administration of PAs (IV vs oral) because PAs are not as clearly restricted by drug in their route of administration from one country to the next, and the route of administration is not reported or inferable from FAERS otherwise.12,30 However, we do think our results show signals that route of administration of PAs may affect sensitization patterns and anaphylaxis. Notably, in comparison to the United States where amoxicillin and amoxicillin-clavulanic acid are available only in oral formulations, other countries use IV formulations of these medications. In particular, both the United Kingdom and France use IV formulations of amoxicillin and amoxicillin-clavulanic acid and these 2 countries contributed the most anaphylaxis reports to amoxicillin and amoxicillin-clavulanic acid in FAERS (Figures 2 and 3). It is possible that IV amoxicillin administration in these countries may more effectively sensitize patients to IgE-mediated allergy or alternatively that oral medications lead to more tolerance. Both of these explanations could add insight into the differential reporting patterns of anaphylaxis in our current study and also differences in skin testing sensitization patterns seen in other studies.1,21–28 Alternatively, it is also possible that our findings reflect the fact that IV administration can result in more severe and noteworthy reactions, leading to higher reporting of anaphylaxis.1 Lastly, another possibility is that these findings only reflect regional utilization patterns of different drugs, without differential patterns in sensitization and allergy. This remains a possibility because the FAERS database does not contain drug administration data; therefore, we cannot calculate the true incidence of anaphylaxis reports on the basis of a denominator of doses of drug given. Future epidemiologic studies are needed to investigate this further, including more studies assessing the true incidence of penicillin allergy and anaphylaxis using national databases with different formularies. Future mechanistic studies would also help elucidate whether IV administration of PAs may contribute to the development of IgE-mediated allergy.
Our results are also elucidating when paired with findings from Liang et al’s aforementioned large population-based observational study using electronic medical record data to look at the incidence of anaphylaxis to PAs.4 Although these data cannot be directly compared given the contrasting study methodologies, they can be considered complementary. For example, Liang et al found a higher risk of anaphylaxis with parenteral PAs compared with oral PAs: anaphylaxis occurred in 1 of 123,792 oral PA courses versus 1 of 255,320 parenteral PA courses (P < .001). This supports that IV administration may present a higher risk of anaphylaxis, whether from sensitization or from increased bioavailability leading to a faster onset, more severe reaction.1 Another noteworthy finding from Liang et al’s study was that among the studied PAs, dicloxacillin had the highest incidence of anaphylaxis (0.005% compared with <0.001% for amoxillin, for example). In comparison, our FAERS data demonstrated low reporting of anaphylaxis to dicloxacillin overall (4 non-US cases total). Another interesting finding in Liang et al’s study was that among the parenteral PAs, piperacillin-tazobactam and ampicillin-sulbactam had higher incidences of anaphylaxis. When examining US cases only (where the route of drug administration is more clear based on the formulary), we observed higher reporting odds of anaphylaxis for piperacillin-tazobactam, but not ampicillin-sulbactam. In summary, although data from these 2 studies are not directly comparable, their complementary findings suggest that future research is needed to determine whether dicloxacillin and ampicillin-sulbactam may be associated with a higher risk of allergy. Taken together, these data reaffirm results from recent skin testing studies suggesting an emerging contemporary presence of selective piperacillin-tazobactam allergy.1,4,29
Lastly, but critically, a widespread feared complication of penicillin allergy is fatal anaphylaxis. Although all antibiotics, including PAs, are among the most common drugs associated with fatal anaphylaxis, few studies to date have been able to examine the incidence of mortality associated with anaphylaxis to PAs, because it is relatively rare and hard to study.31–33 There are significant limitations to our mortality data, including low crude numbers limiting statistical analysis, lack of drug administration data, the possibility of duplicate reports, and the fact that true causality cannot be established (eg, deaths reported by FAERS may be very well due to other conditions, especially among patients who are hospitalized and requiring antibiotics). Nonetheless, we find it noteworthy that the overall number of reported fatalities associated with reported penicillin anaphylaxis was low, especially when considering that PAs are among the most commonly prescribed antibiotics worldwide (Table V). For example, in 11 years of FAERS data in the United States, there were only 4 cases of death reported to be associated with anaphylaxis to amoxicillin-containing medications, and only 18 reports of death associated with reported anaphylaxis to PAs in total. There has been increasing international momentum toward applying risk stratification techniques to assessing penicillin allergy and increasingly using direct oral challenges over the last 5 years. As we await a national/international study examining the true incidence of fatal anaphylaxis with PAs, the very low number of reported cases of anaphylaxis associated with mortality in this large study of FAERS data (including more than 20 million reports of adverse drug reactions) should provide further reassurance that fatal anaphylaxis to PAs is likely to be quite rare.
Study limitations
This study adds value by using previously established pharmacovigilance methods and data from recent years to investigate disproportionality signals in reported anaphylaxis to penicillins in a large reporting database (FAERS).7,12,15 There are limitations to our study, many of which are inherent to using FAERS data. It is important to highlight that the denominator of drug administration instances is unknown, and so it is impossible to calculate the true incidence or relative risk of anaphylaxis. Similarly, without drug administration data, it is currently impossible to know the extent to which differences in reporting odds regionally may reflect regional utilization differences; this could be addressed through studies using national electronic health record databases in the future. In addition, some potentially valuable data were not available in FAERS at the time of this study (eg, route of drug administration). In addition, FAERS data are not stagnant, but refined over time. FAERS data may also be subject to underreporting, over-reporting, and duplication. Lastly, the context of reactions is unknown; we lack clinical histories and testing results that may aid in establishing causality. Future large retrospective studies using clinical data containing total drug administrations and reactions would complement our study and aid in better understanding the incidence of anaphylaxis with PAs and local drug allergy epidemiology patterns.
Conclusion
Our study provides valuable insights regarding penicillin allergy epidemiology through analysis of anaphylaxis events associated with PAs reported to the FDA and is particularly valuable when paired with clinical penicillin allergy testing studies from around the world. Our findings demonstrate that all PAs show higher reporting odds of anaphylaxis (when compared with other medications in the FAERS database). We also show different reporting odds patterns when comparing US to non-US reports. Given complementary clinical drug allergy testing studies, we think that together this implies that the epidemiology of anaphylaxis and allergy to specific PAs may vary regionally, with higher reporting odds of anaphylaxis to amoxicillin-containing medications outside the United States. Similarly, we believe that these data also provide a reminder that allergies to PAs can indeed be selective and side chain—specific. True differences in relative risk would need to be assessed in studies using multinational drug administration data. Future directions that could expand our understanding of anaphylaxis to PAs include large retrospective studies using multinational and large US regional electronic health record databases to assess the incidence of anaphylaxis (with drug administration data), as well as further clinical studies assessing for the prevalence of side chain—selective aminopenicillin and piperacillin allergies among patients with histories concerning for recent IgE-mediated reactions. Future research is also needed to aid in our understanding of how route of administration of medications (eg, IV vs oral) may affect tolerance, sensitization patterns, and anaphylaxis risk.
What is already known about this topic?
Anaphylaxis is a severe and feared result of penicillin allergy, but its epidemiology has been historically difficult to study given its rarity.
What does this article add to our knowledge?
Penicillin antibiotics (PAs) were associated with high reporting odds of anaphylaxis in the Food and Drug Administration Adverse Event Reporting System database. We observed differential reporting of anaphylaxis attributed to specific PAs regionally. Reported fatalities associated with anaphylaxis to PAs were rare.
How does this study impact current management guidelines?
This study reaffirms that PAs are a common cause of reported drug allergy. However, it also showed that anaphylaxis and mortality associated with PAs was rarely reported in the Food and Drug Administration Adverse Event Reporting System database. Lastly, the results suggest that regional epidemiology should be considered during evaluation of allergic reactions to PAs.
Acknowledgments
A.B. has been supported by a Yale Physician Scientist Development Award and by the National Center for Advancing Translational Science (Clinical and Translational Science Award grant no. UL1 TR001863), a component of the National Institutes of Health (NIH), as well as Agency for Research and Quality. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. E.M.M. receives support from a Vanderbilt Faculty Research Scholars grant (NIH K12). C.S. receives support for penicillin allergy—related research from the Agency for Healthcare Research and Quality (grant no. R01HS03234) and receives additional unrelated funding support from the National Institute of Allergy and Infectious Diseases (grant no. U01AI181927) for alpha-gal allergy research and a pilot award for chemotherapy allergy research from the Vanderbilt Ingram Cancer Center/Chic Awareness. E.J. P. is supported by NIH awards (nos. R01HG010863 and R01AI152183), the National Health and Medical Research Council (NHMRC) of Australia (grant no. GNT11234999), the Angela Anderson Fund, and the SJS Research Fund (philanthropic support Vanderbilt University Medical Center), the NIH (grant nos. U01AI154659 and R13AR074889), and the NHMRC of Australia (grant no. GNT2028952) (none of which are related to this work.).
Abbreviations used
- FAERS
Food and Drug Administration Adverse Event Reporting System
- FDA
Food and Drug Administration
- IV
intravenous/intravenously
- PA
penicillin antibiotic
- ROR
reporting odds ratio
- PRR
proportional reporting ratio
Footnotes
Conflicts of interest: E. J. Phillips receives royalties and consulting fees from UpToDate and Uptodate Lexidrug (where she is a Drug Allergy Section Editor and section author) and has received consulting fees from Janssen, Vertex, Verve, Servier, Rapt, and Esperion; is codirector of IIID Pty Ltd, which holds a patent for HLA-B*57:01 testing for abacavir hypersensitivity; and has a patent pending for detection of HLA-A*32:01 in connection with diagnosing drug reaction with eosinophilia and systemic symptoms to vancomycin. For these patents, she does not receive any financial remuneration, and neither are related to the submitted work. C. Stone Jr has served as a consultant for Stallergenes-Greer and Larimar Therapeutics in the past year, unrelated to the submitted work. The authors declare no competing financial interests.
REFERENCES
- 1.Khan DA, Banerji A, Blumenthal KG, Phillips EJ, Solensky R, White AA, et al. Drug allergy: a 2022 practice parameter update. J Allergy Clin Immunol 2022;150:1333–93. [DOI] [PubMed] [Google Scholar]
- 2.Shenoy ES, Macy E, Rowe T, Blumenthal KG. Evaluation and management of penicillin allergy: a review. JAMA 2019;321:188–99. [DOI] [PubMed] [Google Scholar]
- 3.Vogrin S, Blumenthal KG, Trubiano JA. Examining the utility of penicillin allergy clinical decision rules: how low will you go? J Allergy Clin Immunol Pract 2024;12:2994–5. [DOI] [PubMed] [Google Scholar]
- 4.Liang EH, Chen LH, Macy E. Adverse reactions associated with penicillins, carbapenems, monobactams, and clindamycin: a retrospective population-based study. J Allergy Clin Immunol Pract 2020;8:1302–1313.e2. [DOI] [PubMed] [Google Scholar]
- 5.International Rheumatic Fever Study Group. Allergic reactions to long-term benzathine penicillin prophylaxis for rheumatic fever. Lancet 1991;337:1308–10. [PubMed] [Google Scholar]
- 6.Lee P, Shanson D. Results of a UK survey of fatal anaphylaxis after oral amoxicillin. J Antimicrob Chemother 2007;60:1172–3. [DOI] [PubMed] [Google Scholar]
- 7.Roger JY, Krantz MS, Phillips EJ, Stone CA Jr. Emerging causes of drug-induced anaphylaxis: a review of anaphylaxis-associated reports in the FDA Adverse Event Reporting System (FAERS). J Allergy Clin Immunol Pract 2021;9:819–829.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jerschow E, Lin RY, Scaperotti MM, McGinn AP. Fatal anaphylaxis in the United States, 1999–2010: temporal patterns and demographic associations. J Allergy Clin Immunol 2014;134:1318–1328.e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Feinberg SM, Feinberg AR, Moran CF. Penicillin anaphylaxis, nonfatal and fatal reactions. J Am Med Assoc 1953;152:114–9. [DOI] [PubMed] [Google Scholar]
- 10.Su C, Belmont A, Liao J, Kuster JK, Trubiano JA, Kwah JH. Evaluating the PEN-FAST Clinical Decision-making Tool to enhance penicillin allergy delabeling. JAMA Intern Med 2023;183:883–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Trubiano JA, Vogrin S, Chua KYL, Bourke J, Yun J, Douglas A, et al. Development and validation of a penicillin allergy clinical decision rule. JAMA Intern Med 2020;180:745–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Shah D, Rukasin C, Wang C, Phillips E, Stone C. Unraveling cephalosporin-associated anaphylaxis: a comprehensive analysis using FDA Adverse Event Reporting System data. J Allergy Clin Immunol Pract 2024;12:1359–61. [DOI] [PubMed] [Google Scholar]
- 13.Yin Y, Shu Y, Zhu J, Li F, Li J. A real-world pharmacovigilance study of FDA Adverse Event Reporting System (FAERS) events for osimertinib. Sci Rep 2022;12:19555. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bluestein SB, Yu R, Stone C, Phillips EJ. Reporting of drug reaction with eosinophilia and systemic symptoms from 2002 to 2019 in the US Food and Drug Administration Adverse Event Reporting System. J Allergy Clin Immunol Pract 2021;9:3208–3211.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Huang H, Li L, Wu M, Liu Z, Zhao Y, Peng J, et al. Antibiotics and antibiotic-associated diarrhea: a real-world disproportionality study of the FDA adverse event reporting system from 2004 to 2022. BMC Pharmacol Toxicol 2023;24:73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhou C, Peng S, Lin A, Jiang A, Peng Y, Gu T, et al. Psychiatric disorders associated with immune checkpoint inhibitors: a pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database. EClinicalMedicine 2023;59:101967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453–7. [DOI] [PubMed] [Google Scholar]
- 18.Mukherjee EM, Park D, Martin-Pozo M, Phillips EJ. Characterization of demographics, drug latency, and mortality of severe cutaneous adverse reactions in an FDA pharmacovigilance database. medRxiv 2025. 2025.03.05.25323441. [Google Scholar]
- 19.Stone CA Jr, Trubiano JA, Phillips EJ. Testing strategies and predictors for evaluating immediate and delayed reactions to cephalosporins. J Allergy Clin Immunol Pract 2021;9:435–444.e13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Doña I, Guidolin L, Bogas G, Olivieri E, Labella M, Schiappoli M, et al. Resensitization in suspected penicillin allergy. Allergy 2023;78:214–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Romano A, Valluzzi RL, Caruso C, Zaffiro A, Quaratino D, Gaeta F. Tolerability of cefazolin and ceftibuten in patients with IgE-mediated aminopenicillin allergy. J Allergy Clin Immunol Pract 2020;8:1989–1993.e2. [DOI] [PubMed] [Google Scholar]
- 22.Romano A, Valluzzi RL, Gaeta F, Caruso C, Zaffiro A, Quaratino D, et al. The combined use of chronological and morphological criteria in the evaluation of immediate penicillin reactions: evidence from a large study. J Allergy Clin Immunol Pract 2022;10:3238–3248.e2. [DOI] [PubMed] [Google Scholar]
- 23.Romano A, Caubet J-C. Antibiotic allergies in children and adults: from clinical symptoms to skin testing diagnosis. J Allergy Clin Immunol Pract 2014;2:3–12. [DOI] [PubMed] [Google Scholar]
- 24.Gadde J, Spence M, Wheeler B, Adkinson NF. Clinical experience with penicillin skin testing in a large inner-city STD clinic. JAMA 1993;270:2456–63. [PubMed] [Google Scholar]
- 25.Solensky R, Jacobs J, Lester M, Lieberman P, McCafferty F, Nilsson T, et al. Penicillin allergy evaluation: a prospective, multicenter, open-label evaluation of a comprehensive penicillin skin test kit. J Allergy Clin Immunol Pract 2019;7:1876–1885.e3. [DOI] [PubMed] [Google Scholar]
- 26.Belmont AP, Su C, Kwah J. Cephalosporins can be safely administered to patients with verified penicillin allergy. J Allergy Clin Immunol Pract 2023;11:2933–2935.e2. [DOI] [PubMed] [Google Scholar]
- 27.Bousquet P, Co-Minh H, Arnoux B, Daures J, Demoly P. Importance of mixture of minor determinants and benzylpenicilloyl poly-l-lysine skin testing in the diagnosis of β-lactam allergy. J Allergy Clin Immunol 2005;115:1314–6. [DOI] [PubMed] [Google Scholar]
- 28.Matheu V, Pérez E, González R, Poza P, de la Torre F, Sánchez-Machín I, et al. Assessment of a new brand of determinants for skin testing in a large group of patients with suspected beta-lactam allergy. J Investig Allergol Clin Immunol 2007;17:257. [PubMed] [Google Scholar]
- 29.Casimir-Brown RS, Kennard L, Kayode OS, Siew LQ, Makris M, Tsilochristou O, et al. Piperacillin-tazobactam hypersensitivity: a large, multicenter analysis. J Allergy Clin Immunol Pract 2021;9:2001–9. [DOI] [PubMed] [Google Scholar]
- 30.Park B, Pirmohamed M, Kitteringham N. Idiosyncratic drug reactions: a mechanistic evaluation of risk factors. Br J Clin Pharmacol 1992;34:377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Idsoe O, Guthe T, Willcox R, De Weck A. Nature and extent of penicillin side-reactions, with particular reference to fatalities from anaphylactic shock. Bull World Health Organ 1968;38:159. [PMC free article] [PubMed] [Google Scholar]
- 32.Blumenthal KG, Smith LR, Mann JTS, Salciccioli I, Accarino JJ, Shah RJ, et al. Reaction risk to direct penicillin challenges: a systematic review and meta-analysis. JAMA Intern Med 2024;184:1374–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Turner PJ, Jerschow E, Umasunthar T, Lin R, Campbell DE, Boyle RJ. Fatal anaphylaxis: mortality rate and risk factors. J Allergy Clin Immunol Pract 2017;5:1169–78. [DOI] [PMC free article] [PubMed] [Google Scholar]