Evaluation of cephalosporin allergy: Survey of drug allergy experts

Abstract

Background

Since the publication of the 2022 Drug Allergy Practice Parameters (DAPP) of the American Academy of Allergy, Asthma & Immunology (AAAAI) and American College of Allergy, Asthma & Immunology (ACAAI), it is unclear the extent to which the simplified and risk-stratified evaluation of cephalosporin allergy has been incorporated into allergy practice.

Objective

We aimed to assess current cephalosporin allergy testing practices using real case examples.

Methods

An 18-question REDCap survey was sent to the 136 members of the Adverse Reactions to Drugs, Biologics and Latex (ARDBL) Committee of the AAAAI between February and April 2023.

Results

Forty-six (33.8%) ARDBL members completed the survey after 3 email attempts. Most practiced in the United States (32, 69.6%), 6 (13.0%) in Canada, and the rest in Europe and Asia. Almost half (47.7%) reported that the 2022 DAPP had increased their use of direct oral challenge, and 91% would prescribe cephalosporins in the setting of low-risk penicillin allergy history without testing. For low-risk cephalosporin reactions, 68% would perform a direct oral challenge with the culprit drug. In severe immediate penicillin reactions, 23% would evaluate with penicillin skin test before assessing cephalosporin allergy. For cephalosporin-related anaphylaxis, 48% would perform cephalosporin-based tests. For perioperative anaphylaxis with cefazolin, 57% would perform cephalosporin-based tests. For positive skin test result to cefazolin, 79% chose to avoid the culprit drug with follow-up oral challenge to a structurally dissimilar cephalosporin.

Conclusion

Increased uptake of direct oral challenge represents the initial impact of the 2022 DAPP. However, there is significant variation in testing practices of cephalosporin allergy even among drug allergy experts, reflecting a need for a firmer evidence base to guide consensus around testing for higher-risk reactions.

Cephalosporin antibiotics are commonly prescribed for a wide range of infections as a result of their effectiveness and broad-spectrum activity. These drugs rarely cause diverse immunologic reaction phenotypes across different age groups, including IgE-mediated reactions, benign T-cell–mediated exanthems as well as serum sickness–like, single organ damage, and severe cutaneous adverse reactions. At the same time, as a result of their common use, a cephalosporin allergy label is relevant for 1.3% to 1.7% of the general population in the United States.¹

Despite major progress in understanding more about reactions to cephalosporins, specific and validated testing strategies of cephalosporin allergy evaluation are less well defined.² In the last few years, there is evidence that immediate allergic reactions to cephalosporins as well as their cross-reactivity with penicillin appears to be largely related to antigenic responses to R1 side chains rather than β-lactam ring cross-reactivity or R2 side chains.^3,4 Data on cross-reactivity among delayed reaction types are not as well characterized.

With improvements in our understanding of cross-reactivity and reaction risk assessment, the joint 2022 Drug Allergy Practice Parameters (DAPP) of the American Academy of Allergy, Asthma & Immunology (AAAAI) and American College of Allergy, Asthma & Immunology (ACAAI) now advocate a simplified and risk-stratified evaluation for cephalosporin allergy.

For immediate reactions to cephalosporins, the 2022 practice parameter suggests stratifying patients on the basis of potential IgE-mediated anaphylactic reactions versus nonanaphylactic reactions. For those patients with nonanaphylactic cephalosporin allergy, a direct challenge should be performed for a cephalosporin with dissimilar side chains to determine tolerance. In contrast, for administration of cephalosporins with similar side chains and for the less common anaphylactic reaction history, a negative cephalosporin skin test result to a parenteral cephalosporin is required before challenge to determine tolerance.³ Cephalosporin skin tests have not been validated and have lower sensitivities than penicillin, so ideally, negative results should always be followed by graded challenges.¹

On the subject of cross-reactivity between penicillins and cephalosporins, the practice parameter suggests again stratifying patients on the basis of anaphylactic versus nonanaphylactic histories, as well as determining whether the penicillin allergy is confirmed or unconfirmed. For patients with a history of an unconfirmed nonanaphylactic penicillin allergy, any cephalosporin can be routinely administered without testing or additional precautions. In contrast, for those rare patients with a history of anaphylaxis to penicillin, a non–cross-reactive cephalosporin can be administered routinely without prior testing.³

Despite the simplified approach to cephalosporin evaluation, the extent to which this approach is incorporated into allergy practice remains unclear. Therefore, we surveyed the members of the Adverse Reactions to Drugs, Biologics, and Latex (ARDBL) Committee of the AAAAI between February and April 2023 to assess current practices of cephalosporin allergy testing using real case examples.

Methods

We conducted a survey on current allergy clinic practices in relation to evaluation of cephalosporin allergy through the special-interest ARDBL Committee. At the time of our research, the committee included 136 members, who met both virtually and in person on average 4 times per year (including in person at the annual AAAAI meeting) to discuss and plan priorities in research, education, and clinical practice in drug allergy.

An 18-question descriptive survey in REDCap was sent to ARDBL Committee members, with 3 survey attempts made between February and April 2023. The survey included questions on demographics (country of practice, years in practice, types of drugs and reaction seen, availability of testing, etc) as well as current practices regarding cephalosporin allergy tests. In order to evaluate the latter, we used real case examples after each case presentation (high or low history of receipt of penicillin or cephalosporin), the survey participants were asked to answer the following question: “How do you evaluate this patient for cephalosporin allergy?” Multiple test options were offered, and more than one option could be chosen.

Five case scenarios of patients with previous reactions to β-lactams were presented to survey participants for their opinions on how they would evaluate and manage each case. The cases were as follows:

• A.Low-risk history with penicillin.
• B.Low-risk history with cephalosporin (cephalexin).
• C.High-risk penicillin history.
• D.High-risk history with cephalosporin (ceftriaxone).
• E.High-risk history with cephalosporin (cefazolin).

Participants were asked to choose the preferred test for each case, specifically focused on answering the questions of future cephalosporin receipt and testing. Several answer choices for tests were offered—for example, penicillin skin test, penicillin oral challenge, cephalosporin skin test, or oral challenge with a dissimilar cephalosporin. Of note, we deliberately chose the term “oral challenge” in order to allow the choice of this test option both with or without a preceding skin test. The term “direct oral challenge” (DOC) was only used in retrospect, when the chosen evaluation options did not include skin test before oral challenge.

We studied 2 cases of anaphylaxis to cephalosporins because we had a specific interest in the evaluation of cefazolin reactions and whether this drug is recognized by drug allergy experts as being primarily non–cross-reactive to both penicillins and cephalosporins. The practice parameter’s algorithm does not recommend a different evaluation for cefazolin, but it does mention the unique characteristics of this drug in its discussion of R1 side chain and cross-reactivity.³

We present the results in such a way as to visually illustrate the distribution of answer combinations, with answers aligning with the practice parameter marked to help visually recognize trends of adherence to the guidelines.

Results

Survey respondents

The demographics of the survey respondents are shown in Table I. Most of the respondents were physicians practicing in the United States (32, 69.6%), along with 6 (13.0%) in Canada, 1 in Europe, and the remaining 7 (15.2%) from countries in Asia. All physicians who took part in the survey identified themselves as seeing patients with drug allergies. Among the most common allergies were antibiotics (43, 95.6%), nonsteroidal anti-inflammatory drugs (41, 91.1%), contrast media (37, 82.2%), and biologics (35, 77.8%). Less common drugs included chemotherapy (29, 64.4%) and miscellaneous drugs (23, 51.1%) such as iron, steroids, and general and local anesthetics. All but 2 (95.7%) of 44 experts saw patients for evaluation of cephalosporin allergy.

Table I.

Characteristics of drug allergy experts

Characteristic	No. (%)
Type of practice
Private allergy practice—solo	2 (4.4)
Private allergy practice—group	2 (4.4)
Group owned by medical center	1 (2.2)
Academic medical center/clinic	40 (88.9)
Multispecialty practice	3 (6.7)
Practice location
United States	32 (69.6)
Canada	6 (13.0)
Europe	1 (2.2)
Other	7 (15.2)
Experience in drug allergy
1-5 years	17 (37.0)
5-10 years	8 (17.4)
10-15 years	8 (17.4)
>15 years	13 (28.3)
Main sources of drug allergy knowledge
Consensus guidelines (practice parameters)	43 (95.6)
Other published articles	39 (86.7)
Conferences and meetings	33 (73.3)
Fellowship training	27 (60.0)
Personal experience	29 (64.4)
Mentors or other experienced colleagues	27 (60.0)
Other	2 (4.4)
Test preparation
Access to pharmacy support	36 (81.8)
Self-obtained and prepared in clinic	13 (29.5)
Both	1 (2.3)
Availability to perform immediate and delayed drug reaction test
Immediate	10 (21.7)
Delayed	0
Both immediate and delayed reactions	36 (78.3)
Type of test performed
Prick test for immediate hypersensitivity	42 (93.3)
Intradermal test for immediate hypersensitivity	42 (93.3)
Prick test for delayed hypersensitivity	10 (22.2)
Intradermal test for delayed hypersensitivity	27 (60.0)
Patch test for delayed hypersensitivity	17 (37.8)
Single-dose challenge for immediate hypersensitivity	33 (73.3)
Two-step challenge for immediate hypersensitivity	39 (86.7)
Single-dose challenge for delayed hypersensitivity	27 (60.0)
Two-step challenge for delayed hypersensitivity	18 (40.0)
Multiple-dose–multiple-day challenge for delayed hypersensitivity	17 (37.8)
Other	2 (4.4)
Type of drug allergy categories seen in practice
Penicillin alone	33 (73.3)
Multiple antibiotics	43 (95.6)
Biologics	35 (77.8)
Contrast media	37 (82.2)
NSAIDs	41 (91.1)
Chemotherapy	29 (64.4)
Other (iron, etc)	23 (51.1)
Cephalosporin hypersensitivity evaluation in the practice
Yes	44 (95.7)
No, but plan to add it to my practice over the next year	1 (2.2)
No, and don’t plan to add it to my practice	1 (2.2)
Most common immediate type reaction to cephalosporin
Cefazolin	24 (53.3)
Ceftriaxone	3 (6.7)
Cephalexin	18 (40.0)
Time from initial reaction to evaluation
1-4 weeks	2 (4.5)
4-8 weeks	12 (27.3)
2-6 months	11 (25.0)
6-12 months	6 (13.6)
1-5 years	3 (6.8)
>5 years	10 (22.7)

NSAID, Nonsteroidal anti-inflammatory drug.

Forty members (88.9%) practiced in an academic medical center, 2 (4.4%) were part of a solo practice, 2 (4.4%) were part of a group allergy practice, and 3 (6.7%) were part of a multispecialty practice.

The experience of the participants in the field of drug allergy was 1-5 years (17, 37.0%), 5-10 years (8, 17.4%), 10-15 years (8, 17.4%), and >15 years (13, 28.3%).

The main sources of drug allergy knowledge reported by the survey respondents were consensus guidelines (practice parameters) (43, 95.6%), other published articles (39, 86.7%), conferences and meetings (33, 73.3%), fellowship training (27, 60.0%), personal experience (29, 64.4%), mentors or other experienced colleagues (27, 60.0%) and other (2, 4.4%).

Thirty-six physicians (78.3%) reported having a capacity to test for both delayed and immediate reaction to cephalosporins, and 10 (21.7%) reported that they only test immediate reactions.

The second section of the survey was only available to participants who stated that they assess patients with cephalosporin allergy in their practice or if they planned to do so in the next year (44, 95.7%). The most common immediate cephalosporin reaction that participants reported evaluating was cefazolin (24, 53.3%), followed by cephalexin (18, 40.0%) and then ceftriaxone (3, 6.7%).

Survey results

Participant responses to how they would evaluate and test each of the 5 case scenarios (scenarios A-E) with a focus on cephalosporin use are shown in Fig 1, with answers aligning with current recommendations for testing in the practice parameters marked with an orange bracket.

Fig 1.

Survey results for expert choice on tests for each case.

For scenario A, the practice parameters suggest that a patient reporting a low-risk penicillin allergy can receive a cephalosporin normally, without a preceding cephalosporin test. Some form of this response was selected by 91% of survey respondents, who focused on prescribing the cephalosporin, delabeling the low-risk penicillin allergy, or both.

For scenario B, the practice parameters suggest that a patient with a low-risk allergy to a cephalosporin should be evaluated by skin test before prescribing a structurally similar cephalosporin. In the survey, 12% chose to perform skin test before challenge, and 68% recommended direct challenge with the culprit drug.

For scenario C, the practice parameters recommend normal administration of a dissimilar cephalosporin, as well as a skin test–based approach for a structurally similar cephalosporin, for patients with an anaphylactic history to a penicillin. Our survey results showed that 23% elected to perform a penicillin-based skin test and focus only on proving or disproving the penicillin allergy, which aligns with practice parameters’ recommendations.

For scenario D, for a patient with a high-risk cephalosporin history, the practice parameters suggest cephalosporin skin test as the evaluation approach. Forty-eight percent of the experts in our survey chose answers that aligned with this approach.

For scenario E, for patients with another high-risk history of a cephalosporin, specifically cefazolin, a higher number of participants, 57%, chose cephalosporin skin test as their evaluation.

Following scenario E, we specifically asked a question about the recommended steps in cephalosporin use when the skin test results are positive to cefazolin. For this scenario, most experts (34, 79.1%) recommended avoiding cefazolin and conducting an oral challenge with a structurally dissimilar cephalosporin, versus 9 experts (20.9%) who recommended avoiding cefazolin alone, without the need to continue to oral challenge.

We attempted to perform a post hoc analysis to look for a correlation between the experts’ time in practice and their adherence to the guidelines, but we found no such correlation in any of the case scenarios.

The survey’s last section asked about the impact of the 2022 practice parameter on the survey respondents’ individual practices.

Twenty-one participants (47.7%) agreed that the 2022 practice parameters had changed their approach specifically for cephalosporin allergy testing. Among the 21 who answered “yes” to having changed their practice after the updated drug allergy practice parameters were reported qualitative changes in their practice of doing more DOCs (7 experts), more single-dose challenges (2 reports), and fewer skin tests (6 reports), as well as being less concerned about penicillin and cephalosporin cross-reactivity.

Discussion

The results of our survey suggest that even among experienced, interested experts in drug allergy, significant variability in practice and incorporation of recent drug allergy practice parameters are present when evaluating cephalosporin allergy. Most agree that DOC is a reasonable approach to evaluate both low-risk penicillin and cephalosporin allergy labels. An abundance of retrospective and prospective studies support DOCs to evaluation for low-risk penicillin allergy;^5,6 recent evidence of a similar approach is emerging for cephalosporin allergy, which has led to updated practice parameters regarding cephalosporin evaluation.

A study by Sillcox et al,⁷ performed a DOC in 136 pediatric patients with a history of nonvesicular skin-limited symptoms while being treated with cephalosporins, concluded that this approach is a safe and effective method for delabeling cephalosporin allergy in children. Koo et al⁸ validated a history-based risk stratification of cephalosporin allergy and suggested it to be a feasible adaptation of an existing validated criteria for penicillin allergy. Although higher-level evidence is needed, particularly in adults, there already appears to be a fair level of comfort among drug allergy experts in using DOC to evaluate low-risk cephalosporin reactions.⁶ This is important because there is no good skin test strategy for reactions to aminocephalosporins, which are only available as oral preparations. Although the drug allergy practice parameters currently recommend skin test before oral challenge for low-risk cephalosporin reactions, the surveyed experts showed a possible tendency toward being liberal and offering a DOC, similar to current guidance for low-risk penicillin allergy labels.

When evaluating a patient with a high-risk penicillin allergy, many experts chose to include both penicillin and cephalosporin skin tests, as opposed to just performing a penicillin-based skin test, as suggested by the practice parameter. Similarly, for a high-risk cephalosporin history, more physicians recommended a penicillin skin test in addition to a cephalosporin skin test, even when evaluating whether the patient can receive cephalosporins, while the practice parameter does not suggest the need to evaluate penicillins.

We were especially interested in the approach to cefazolin allergy among drug allergy experts. Cefazolin is a first-generation cephalosporin broadly used for perioperative prophylaxis.⁹ Reviewing the chemical structures and immunogenic components of cefazolin shown that it has 2 side chains (R1 and R2) whose structures do not match those of any other β-lactams in use in the United States, including penicillin, which suggests that the true rate of cross-reactivity of cefazolin to other β-lactam antibiotics may actually be close to zero.¹⁰ Several studies suggest that patients with past immediate reactions to cefazolin often tolerate all other cephalosporins and other β-lactams.^2,3,11 A 2021 study demonstrated that in a series of 452 patients labeled as having penicillin allergy who underwent allergy testing, none had a positive skin test result when tested specifically for cefazolin or ceftriaxone.² In a systematic review and meta-analysis from 2021 looking at dual allergy to penicillin and cefazolin, less than 1% of nonconfirmed penicillin allergy also had the label of cefazolin allergy, suggesting that most patients with a history of penicillin allergy may safely be prescribed cefazolin.¹²

In our study, 48% of experts chose a cephalosporin-only skin test in the case of anaphylaxis to ceftriaxone (an antibiotic that shares side chains with several other cephalosporins, such as cefpodoxime, cefotaxime, and cefepime) versus 57% who chose a cephalosporin-only skin test in cases of immediate reactions to cefazolin. However, when specifically asked what the preferred step is after a positive skin test result to cefazolin, most experts (34, 79.1%) recommended avoiding cefazolin and conducting an oral challenge with structurally dissimilar cephalosporin, versus 9 experts (20.9%) who recommended avoiding cefazolin only, without a challenge with a dissimilar cephalosporin.

The main limitations of this study are the small sample size, low response rate of 33.8% among all ARDBL members, and lack of independent evaluation of cephalosporin allergy in children. However, large surveys typically show response rates of <20%, even with multiple callouts.¹³ A Canadian study that looked at response rates to web-based surveys among physician specialists demonstrated an average response rate of 35%,¹⁴ which is similar to our response rate. The other limitation is the relatively short period of time between the publication of the 2022 practice parameter (online publication in September 2022) and performance of the survey (February-April 2023). Administering the survey later, after the practice parameter’s official publication, might have shown different results, with more experts implementing the guidelines.

Another potential limitation is that the majority of responders were from academic medical centers, with only a small representation of drug allergy experts from the community. However, the selection of ARDBL members for the survey was intentional: they represent the top experts in drug allergy and are the first practitioners to be updated on the new guidelines. We selected them for our survey to see if there was general agreement among this group of experts on the subject of cephalosporin allergy evaluation, especially after the 2022 publication of the DAPP.

The variations among drug allergy experts regarding high-risk reactions with cephalosporins in particular suggest that there is currently low consensus even among experts, meaning that community members will not be expected to have better agreement. However, as a result of the limited access to reagents for drug allergy testing, in the community, high-risk cephalosporin allergy is usually not evaluated, with such patients referred to an academic medical center.

However, for the same reason of limited resources, if further prospective studies provide an evidence base for low-risk cephalosporin allergy evaluation through DOCs, drug allergy providers in the community may actually perform DOCs with cephalosporins.

In support of the idea that drug allergy practice in the United States is rapidly evolving, with experts receptive to the evolving evidence and ready to change their practice guidelines, most reported that they performed more DOCs, more single-dose challenges, and fewer skin tests, and they also affirmed that the updated DAPP had already influenced their practice.

Conclusions

Drug challenge remains the reference standard to remove a drug allergy label, and in low-risk cases, performing skin tests before drug challenge may not be that helpful. Increased uptake of DOC, as shown in our survey, may represent an impact of the 2022 DAPP and associated guideline recommendation changes from the 2010 version. However, we observed significant variation in test practices of cephalosporin allergy among drug allergy experts when evaluating high-risk reactions, with variations among the experts as well as variations from the recommended DAPP pathways. This likely reflects a need for a better evidence base to guide the consensus on testing higher-risk reactions in clinical practice.

Disclosure statement

Disclosure of potential conflict of interest: E. J. Phillips reports grants and funding from the National Institutes of Health (R01HG010863, R01AI152183, U01AI154659, R13AR078623, UAI109565) and from the National Health and Medical Research Council of Australia; receives royalties from UpToDate and consulting fees from Janssen, Vertex, Biocryst, Regeneron, and Verve; and is codirector of IIID Pty Ltd, which holds a patent for HLA-B∗57:01 testing for abacavir hypersensitivity, and which has a patent pending for “Detection of Human Leukocyte Antigen-A∗32:01 in Connection With Diagnosing Drug Reaction With Eosinophilia and Systemic Symptoms,” without any financial remuneration and not directly related to the report. C. A. Stone Jr received funding from an AAAAI Foundation Faculty Development Award that directly supported this research. The other author declares that she has no relevant conflicts of interest.