β-Lactam allergy testing and delabeling-Experiences and lessons from Singapore

Reaction Risk to Direct Penicillin Challenges: A Systematic Review and Meta-Analysis

Importance

While direct penicillin challenges might support the expansion of penicillin allergy delabeling efforts, the perceived risk of reactions remains a key barrier.

Objective

To evaluate the frequency of reactions to direct penicillin challenges in individuals with penicillin allergy labels and to identify factors associated with such reactions.

Data Sources

Three electronic databases were searched (MEDLINE, Web of Science, and Scopus) from inception to July 19, 2023, for primary studies assessing patients undergoing direct penicillin challenges. Articles were included regardless of publication year, language, status, or definition of allergy risk.

Study Selection

Two reviewers independently selected original studies reporting the frequency of immunologically mediated reactions following a direct penicillin challenge in patients reporting a penicillin allergy.

Data Extraction and Synthesis

Two reviewers independently extracted data and independently assessed the quality of each primary study using a risk-of-bias tool for prevalence studies.

Main Outcomes and Measures

The primary outcome was the frequency of reactions to direct penicillin challenges as calculated using random-effects bayesian meta-analysis of proportions. Secondary outcomes included risk factors for reactions and the frequency of severe reactions.

Results

A total of 56 primary studies involving 9225 participants were included. Among participants, 438 experienced reactions to direct penicillin challenges without prior testing, corresponding to an overall meta-analytic frequency of 3.5% (95% credible interval [CrI], 2.5%-4.6%). Meta-regression analyses revealed that studies performed in North America had lower rates of reaction to direct challenges (odds ratio [OR], 0.36; 95% CrI, 0.20-0.61), while studies performed in children (OR, 3.37; 95% CrI, 1.98-5.98), in outpatients (OR, 2.19; 95% CrI, 1.08-4.75), and with a graded (OR, 3.24; 95% CrI, 1.50-7.06) or prolonged (OR, 5.45; 95% CrI, 2.38-13.28) challenge had higher rates of reaction. Only 5 severe reactions (3 anaphylaxis, 1 fever with rash, and 1 acute kidney injury) were reported, none of which were fatal.

Conclusions and Relevance

This systematic review and meta-analysis found that reactions to direct penicillin challenges are infrequent, with rates comparable to indirect challenges after allergy testing. These findings suggest that direct challenges are safe for incorporation into penicillin allergy evaluation efforts across age groups and clinical settings.

Comparing Skin and Serum Testing to Direct Challenge Outcomes in Children With β-Lactam Allergies

BACKGROUND

There is a scarcity of prospective studies investigating the relative roles of skin prick and intradermal testing, serum specific IgE, and extended oral challenges in diagnosing children with reported β-lactam allergies.

OBJECTIVE

To determine the sensitivity and specificity of skin testing and serum specific IgE in children with β-lactam allergies, with immediate and nonimmediate historic reactions.

METHODS

Four hundred children with parent-reported β-lactam allergies were recruited into an open-label prospective study. Detailed allergy histories were collected. Those with medically observed and documented histories of anaphylaxis, requiring epinephrine, or severe cutaneous adverse reactions were excluded. In total, 380 children underwent all testing modalities and a direct provocation test. Each child was followed up for a minimum of 3 years.

RESULTS

True allergy in children was uncommon; 8.3% reacted to the direct provocation challenge or the 5-day extended oral provocation challenge. Children reporting cephalosporin allergy or a reaction within 1 year were more likely to react to direct provocation testing. The sensitivity, specificity, and positive predictive value of skin testing were 12.5%, 98.8%, and 20.0% for direct challenge outcomes, 4.76%, 99.0%, and 25.0% for extended challenge outcomes, and 6.9%, 99.0%, and 40.0% for both challenges combined, respectively. Follow-up investigations revealed that 5.7% of children had a mild repeat reaction and 2.7% continued to avoid the culprit despite successful delabeling. The relabeling rate for children readmitted to hospital was 15%, with the relabeing being unfounded.

CONCLUSIONS

Genuine β-lactam allergies were rare, with over 90% of children effectively delabeled. Skin and serum specific IgE testing did not aid the diagnosis of β-lactam antibiotic allergy in children, regardless of medical history. Extended oral challenges proved valuable in confirming allergies and boosted parental confidence.

Beta-lactam allergy in patients: an antibiotic stewardship challenge

BACKGROUND

Patients are commonly reported as being allergic to beta-lactam (BL) antibiotics. However, many patients with this reported allergy are able to receive BL treatments because they do not have true allergies. In many cases these are simply intolerances due to side effects reported as an allergy. Delabelling these patients leads to better clinical outcomes, optimal antibiotic usage, decreased bacterial resistance and reduced healthcare costs. Therefore, the aims of this study were to identify incorrectly labelled BL allergies in hospitalised patients and to assess antibiotic use in delabelled patients in order to establish a quality indicator to optimise antimicrobial treatments.

METHODS

A prospective study was conducted in which hospitalised patients treated with antimicrobial drugs and labelled as 'BL-allergic' were identified by clinical pharmacists. An allergist assessed whether patients were suitable candidates for a skin test or oral challenge. The Allergy Service removed 'BL-allergic' labels if negative results were obtained. Delabelled patients were followed up by clinical pharmacists to study the use of BL antibiotics as a result of the delabelling programme.

RESULTS

A total of 176 suspected allergic patients were identified and 91 (51.7%) were tested either by a skin test or oral challenge based on the patient indicators. Seven (16.4%) patients tested were allergic to BL antibiotics, 76 (83.5%) were totally delabelled and eight (0.1%) were partially delabelled. Thirty-two (38.1%) delabelled patients required antibiotic treatment in another inpatient or outpatient setting, of whom 27 (84.3%) patients with a new infectious episode received BL treatments while five (15.7%) continued to receive antimicrobial treatments without BL.

CONCLUSION

After the implementation of a protocol to detect incorrect BL allergy labels, 83.5% of the patients in this cohort were completely delabelled. This shows that there is a clear opportunity to optimise the use of antibiotics by delabelling 'BL-allergic' patients.

EAACI/ENDA position paper on drug provocation testing

In drug hypersensitivity, drug provocation testing (DPT), also called drug challenge, is the gold standard for investigation. In recent years, risk stratification has become an important tool for adjusting the diagnostic strategy to the perceived risk, whilst still maintaining a high level of safety for the patient. Skin tests are recommended before DPT but may be omitted in low-risk patients. The task force suggests a strict definition of such low-risk patients in children and adults. Based on experience and evidence from studies of allergy to beta-lactam antibiotics, an algorithm on how to adjust DPT to the risk, and when to omit skin tests before DPT, is presented. For other antibiotics, non-steroidal anti-inflammatory drugs and other drugs, skin tests are poorly validated and DPT is frequently necessary. We recommend performing DPT with chemotherapeutics and biologicals to avoid unnecessary desensitization procedures and DPT with skin tests negative contrast media. We suggest DPT with anesthetics only in highly specialized centers. Specifics of DPT to proton pump inhibitors, anticonvulsants and corticosteroids are discussed. This position paper provides general recommendations and guidance on optimizing use of DPT, whilst balancing benefits with patient safety and optimizing the use of the limited available resources.

Is a Prolonged Drug Provocation Test Better Than a Single-Day Drug Provocation Test? A Systematic Review and Meta-Analysis

BACKGROUND

There is currently no standardized duration of drug provocation test (DPT) for confirming/delabeling beta-lactam hypersensitivity reaction (BL-HSR).

OBJECTIVES

This meta-analysis and systematic review aimed to investigate the added diagnostic value of extended-day over single-day DPT for confirming/delabeling BL-HSR in adults and children.

METHODS

The MEDLINE, EMBASE, Web of Science, and CINAHL online databases were searched from inception to March 15, 2023, for studies that performed extended-day DPT to confirm/delabel BL-HSR. Risk difference and risk ratio were used to compare the proportions of patients with confirmed BL-HSR by single-day or extended-day DPT.

RESULTS

A total of 10,371 DPTs from 42 studies were included. Extended-day DPTs ranged from 2 to 7 days, or as long as index reactions were reported (maximum 10 days). The overall prevalence of confirmed BL-HSR was 6.96% (3.31% during the first-day DPT, and 3.65% during extended-day DPT). Approximately half of the positive reactions during extended-day DPT occurred during the second/third day. The increased detected pool prevalence of confirmed BL-HSR yielded by extended-day DPT was 0.03 (95% CI, 0.02%-0.04%; I2 = 57.69%; P < .001), and the risk ratio of positive reactions between extended-day and single-day DPT was 1.94 (95% CI, 1.62-2.33; I2 = 36.26%; P < .001). The risk difference increased per 1% increase in prevalence of BL-HSR by 0.6% (95% CI, 0.4%-0.7%; P < .001). Twenty-three severe reactions occurred during DPT, and only 2 severe reactions (0.02%) occurred during extended-day DPT. An additional 28 extended-day DPTs were needed to identify 1 mild reaction.

CONCLUSIONS

The increased prevalence of confirmed BL-HSR observed during extended-day DPT could be attributed to the first-day DPT. As a result, our findings do not conclusively support the use of extended-day DPT over single-day DPT. Further studies, incorporating a washout period, are required to comprehensively compare these 2 approaches.

β-Lactam allergy testing and delabeling-Experiences and lessons from Singapore

Background

β‐Lactam allergy is over‐reported and this leads to greater healthcare costs. Allergy testing has inherent risks, yet patients who test negative may continue avoiding β‐lactams.

Objective

To evaluate the safety and diagnostic value of β‐lactams allergy testing locally and usage of antibiotics following negative testing.

Methods

We performed a retrospective medical record review and follow‐up survey of patients who underwent β‐lactam testing between 2010 and 2016 at the National Skin Centre, Singapore.

Results

We reviewed the records of 166 patients, with a total of 173 β‐lactam allergy labels. Eighty (46.2%) labels were to penicillin, 75 (43.1%) to amoxicillin/amoxicillin‐clavulanic acid, 11 (6.4%) to cephalexin, and 5 (2.9%) to others. Skin tests were performed in 142 patients and drug provocation tests (DPTs) in 141 patients. Eleven (6.6%) patients defaulted DPTs after skin testing. Out of 166 patients, 22 (13.3%) patients were proven allergic by either skin tests (16) or DPTs (6). Patients who tested positive had nonsevere reactions.

Out of 155 patients who were conclusively evaluated, 133 (85.8%) were not allergic. Of these patients, 30 (22.6%) used the tested β‐lactam subsequently, with one reporting a mild reaction. Fifty‐one (38.3%) patients were uncontactable or uncertain if they consumed a β‐lactam since testing negative. Fifty‐two (39.1%) patients had no re‐exposure (35 had no indication, 17 were fearful of reactions).

Conclusion

Drug allergy testing was safe and removed inappropriate labels.

Clinical Implication

Allergy testing is efficacious, but fears of subsequent rechallenge should be addressed to maximize the effectiveness of allergy delabeling.