Delayed Penicillin Rash and Later Autoimmunity, With Eyal Kristal, MD

Benign delayed penicillin hypersensitivity in early childhood was associated with a modest increase in later immune-mediated disease in a nationwide Israeli cohort of > 1.3 million children, according to a research letter from investigators at Ben-Gurion University of the Negev.¹

Most children labeled penicillin-allergic after low-risk delayed reactions tolerate subsequent exposure, supporting proactive delabeling.¹˒² Severe T-cell–mediated reactions, including drug reaction with eosinophilia and systemic symptoms (DRESS) and Stevens-Johnson syndrome, have been associated with later immune-mediated disease, but the far more common benign reactions had not been examined in large population-based cohorts.¹

"It's just an epidemiological signal," said Eyal Kristal, MD, from the pediatric allergy and immunology clinic at Saban Children's Hospital, Soroka University Medical Center, in an interview with HCPLive. "For now, there are no clinical steps you need to take due to it."

Kristal said the known link between severe cutaneous reactions and later autoimmunity prompted the team to examine a much more common exposure, the benign rash of childhood, in which immune involvement is also suspected.

Benign Delayed Penicillin Hypersensitivity and Immune-Mediated Disease Risk

Kristal and colleagues conducted a nationwide retrospective population-based cohort study using electronic health records from Clalit Health Services, Israel's largest integrated healthcare organization. The cohort included children aged 0 to 4 years who received ≥ 1 penicillin dispensation between 2000 and 2020, with data available through December 31, 2025.¹

The exposed group comprised children with a documented benign delayed cutaneous reaction occurring 2 to 10 days after penicillin dispensation. Children with urticaria, wheezing, anaphylaxis, epinephrine use, or systemic corticosteroid treatment within 24 hours were excluded, as were those with severe cutaneous adverse reactions.¹ Capping the window at 10 days was intended to exclude DRESS, Kristal explained.

Of 1,375,537 children, 36,531 (2.7%) had a documented benign delayed drug hypersensitivity reaction (DDHR). The composite immune-mediated outcome occurred in 6.2% of children with DDHR vs 5.4% of those without. In the primary double-robust analysis, significant positive associations emerged for celiac disease, vitiligo, autoimmune cytopenias, and Henoch-Schönlein purpura (HSP).¹

Type 1 diabetes showed a modest inverse association (OR, 0.75; 95% CI, 0.58–0.97). A 1:3 propensity score–matched analysis found a similar composite association (OR, 1.08; 95% CI, 1.03–1.13), and the median interval between DDHR and subsequent immune-mediated disease was 7.28 years (interquartile range, 4.37–11.82 years).¹

Detection Bias and Biologic Plausibility in Pediatric Penicillin Allergy

Children with DDHR had greater healthcare utilization before the index event. After additional adjustment for prior primary care visits, emergency department visits, and hospital admissions, celiac disease attenuated to an OR of 1.10 (95% CI, 1.00–1.21). Vitiligo lost significance, and the inverse type 1 diabetes signal disappeared (OR, 0.83; 95% CI, 0.64–1.07).¹

The composite association persisted (OR, 1.08; 95% CI, 1.03–1.13). The strongest remaining signals were autoimmune cytopenias (OR, 1.49; 95% CI, 1.31–1.69) and HSP (OR, 1.48; 95% CI, 1.30–1.68).¹

Kristal acknowledged some bias is inherent to a large retrospective dataset. He noted, however, the multiyear gap between the rash and later diagnosis argues against short-term detection explaining most of the signal. He added the uneven pattern across conditions points away from a single mechanism and toward specific patients with a specific genetic background.

The team hypothesizes the reaction arises from a combination of acute viral illness, beta-lactam exposure, and specific human leukocyte antigen (HLA) susceptibility, driving T-cell hyperactivation. In an analysis not included in the published letter, Kristal said rash diagnoses peaked in December and January, coinciding with viral season. The letter states observational data cannot determine whether the remaining associations reflect shared immunogenetic susceptibility or residual confounding.¹

The investigators wrote the absolute increase was small and should not alter current recommendations for penicillin allergy evaluation and delabeling in children with low-risk histories.¹˒² Confirmation will require prospective studies with validated drug allergy phenotyping. Kristal said such work should measure T-cell activity and cytokine levels during the rash and test whether patients' T cells react against self-antigens tied to specific autoimmune diseases.

“Continue to de-label," Kristal said. "We checked if there is any association to later on autoimmune disease, and it's nothing [to do] with the de-labeling. We don’t think they [necessarily] have [an] allergy to penicillin, so we should continue with the de-labeling.”

References

1. Kristal E, Peles I, Chamudot S, Hazan G. Benign Delayed Penicillin Hypersensitivity in Early Childhood and Subsequent Immune-Mediated Disease. Ann Allergy Asthma Immunol. Published online September 21, 2026. doi:10.1016/j.anai.2026.09.739

2. Blumenthal KG, Peter JG, Trubiano JA, Phillips EJ. Antibiotic allergy. Lancet. 2019;393(10167):183-198. doi:10.1016/S0140-6736(18)32218-9