Zinc supplement: A rare cause of toxic epidermal necrolysis in a pediatric patient

Toxic epidermal necrolysis is a severe cutaneous hypersensitivity reaction associated with high mortality and morbidity.1 In children, drugs are responsible for approximately 60%–90% of Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) cases, with antiepileptics, sulfonamides, and nonsteroidal anti-inflammatory drugs being the most frequent triggers.2 However, reports of SJS/TEN associated with zinc or other nutritional supplements are extremely rare. Although SJS/TEN is most often drug-related, viral and bacterial infections may also act as triggers.2 The lymphocyte transformation test (LTT) assesses in vitro activation of antigen-reactive memory T cells from sensitized individuals. LTT is useful for evaluating the responsible drug for TEN and other nonimmediate type allergic reactions.3 Zinc is an essential mineral that plays a key role in the immune system. Zinc sulfate is frequently used as a supplement, and allergic reactions are exceptionally uncommon. To our knowledge, we report the first pediatric case of toxic epidermal necrolysis related to oral zinc sulfate supplementation. A 5-year-old girl presented with fever and vesiculopapular eruptions for 2 days. The Varicella (chickenpox) was suspected; however, skin lesions progressed and mucosal involvement, in the oral cavity became evident. The patient was previously healthy, with no known chronic diseases, no regular medication use, and no history of allergic or immunologic disorders. A severe cutaneous allergic reaction was considered and intravenous methylprednisolone (1 mg/kg/day) was initiated, along with a single 0.4 g/kg dose of intravenous immunoglobulin (IVIG). A skin biopsy was performed before the commencement of therapy. Rapid evolution to widespread bullous lesions prompted the addition of cyclosporine (CSA; 3 mg/kg/day), and the patient was referred to our pediatric intensive care unit. Upon admission, dermatological examination revealed extensive erythema and epidermal detachment on the face, anterior trunk, and upper extremities, with positive Nikolsky's sign. Ocular, oral, and genital mucosa were involved, and the total body surface area (TBSA) exceeded 30% (Figure 1). According to the SCORTEN scoring system, the patient's score was 4, indicating an estimated mortality risk of 58%. Systemic methylprednisolone dose was increased to 2 mg/kg/day while CSA therapy continued, and IVIG (2 g/kg/dose) was planned over 5 days. Owing to inadequate response, plasmapheresis was initiated, followed by a single dose of etanercept (0.8 mg/kg/dose) after the third session. A total of five consecutive plasmapheresis sessions were performed. Despite possible partial removal of circulating etanercept during the subsequent sessions, fever resolution and cessation of epidermal detachment indicated clinical improvement, suggesting a potential synergistic effect of plasmapheresis and etanercept in the patient's recovery. The patient was discharged on the 15th day with residual postinflammatory hypopigmentation and partial nail bed separations. The skin biopsy demonstrated consistency with the TEN spectrum. When the medication history was reviewed, the patient had received oral clarithromycin daily for 1 week, 1 month prior to admission, and oral zinc sulfate supplementation and a probiotic preparation containing Lactobacillus rhamnosus GG for five consecutive days, completed 10 days before the onset of TEN. She also had intermittent usage of oral vitamin D and omega-3 fatty acid supplements intermittently, 15 days before admission. The ALDEN (Algorithm of Drug Causality in Epidermal Necrolysis) is a tool developed to rapidly identify the culprit drug in SJS/TEN, particularly in cases where patients have been exposed to multiple medications.4 According to the ALDEN criteria, the score was 3 (possible) for zinc sulfate syrup, 0 (unlikely) for the probiotic preparation, and 1 (unlikely) for clarithromycin. Viral serological tests, including Mycoplasma pneumoniae and common viral pathogens, were performed and found to be negative, thereby helping to exclude infectious causes from the differential diagnosis. Four weeks after coming off systemic steroid therapy and planning for withdrawal, LTT was conducted to identify delayed-type hypersensitivity to the culprit agents. The panel included zinc suspension, herbal supplements (including elderberry extract), clarithromycin, and montelukast, ibuprofen, amoxicillin-clavulanate had been consumed within the 8 weeks preceding symptom onset. A significant proliferative response stimulation index (SI) was detected >3 exclusively for the zinc suspension. The other agents were negative or non-reactive (Table 1). One year after the reaction, patch testing was performed with all agents except zinc sulfate and yielded negative results. Patch testing with zinc sulfate could not be conducted because the patient's family did not consent, as it was considered potentially risky. Severe cutaneous drug hypersensitivity reactions are rare but associated with high mortality and morbidity.1 Patients exhibiting acute onset of at least two mucosal involvement and maculae, target-like bullae or erosions, positive Nikolsky sign with skin involvement of more than 30% of TBSA is defined as TEN. The mortality rate of TEN is between 25 and 35%. The latency between drug exposure and symptom onset typically ranges from 4 to 24 days, with a maximum period of 8 weeks.5 Withdrawal of culprit medication and supportive multidisciplinary care are critically important. Although the role of systemic corticosteroids in the treatment of TEN is controversial, they are useful for preventing inflammation in the early stages of the disease.6 IVIG may reduce keratinocyte damage by blocking the FasL/Fas interaction. CSA treatment is recommended to be preferred in the early period to prevent progressive epidermal necrolysis.7 Plasmapheresis has been evaluated as a treatment option in case report series. Plasmapheresis can remove circulating drugs and eliminates drug metabolites and inflammatory mediators such as cytokines in TEN. T-cell-mediated disease with CD8+ cells causes keratinocyte death with the contribution of granulysin, soluble Fas ligand, and tumor necrosis factor-alpha (TNF-a). Etanercept is a monoclonal antibody that functions as a TNF-α antagonist by fusing to the extracellular binding domain of the human TNF receptor 2.8 Few case reports have described the use of etanercept in pediatric patients resistant to the treatments mentioned above. Previous case reports in children have mainly involved anticonvulsants such as lamotrigine and carbamazepine, or antibiotics including amoxicillin and sulfonamides, whereas supplement-related TEN remains exceptional. Several reports in the literature have described SJS/TEN cases triggered by nondrug supplements or mineral preparations. For instance, Fogh reported the development of TEN in a female patient following the intake of high-dose vitamins, minerals, and natural medications.9 Similarly, Al Bahri et al. described a case of SJS/TEN overlap induced by collagen supplementation.10 The exact mechanism by which zinc triggers TEN remains unclear. However, proposed mechanisms include immune dysregulation through altered T-cell responses, cytokine imbalance, and oxidative stress–induced keratinocyte apoptosis. In TEN, cytotoxic CD8+ T cells play a central role in epidermal damage through mediators such as granulysin, soluble Fas ligand, perforin, and TNF-α, leading to widespread keratinocyte death. Zinc excess or hypersensitivity to zinc-containing formulations or excipients may enhance these cytotoxic pathways or promote aberrant T-cell activation in genetically susceptible individuals, thereby facilitating severe epidermal injury.1, 6, 8 Diagnostic tests for delayed drug hypersensitivity are restricted. In vivo tests like epicutaneous patches are the most readily available, must be prepared with the suspected drug in an appropriate concentration in order to yield accurate results. A positive result confirms the involvement of the drug, but the negative predictive value is inadequate. Drug provocation test (DPT) is still considered the gold standard for diagnosing. In most delayed reactions, this option is not possible due to the patient's risk of reaction and the lack of standardization of the challenged drugs. In recent years LTT has been used to diagnose delayed drug-induced hypersensitivity reactions by detecting the proliferation of drug-specific memory T cells.11 In vitro, T cells that encounter the drug and begin to proliferate incorporate tritiated thymidine into their DNA, using three components for each patient: a negative control, a sample containing the drug, and a positive control containing tetanus toxoid. The uptake of thymidine by the cells is measured and the amount taken up per minute is expressed as “counts per minute (cpm).”12 The results indicate the stimulation index (SI) in response to the tested drug at three different dilution levels. The SI represents the ratio of thymidine uptake (measured in cpm) in drug-stimulated cells to that in unstimulated (blank) controls. For non-β-lactam drugs, a value of >3 is positive, 2–3 is borderline or equivocal, and <2 is negative.12 The test should be performed 4–8 weeks after the acute allergic reaction. The sensitivity and specificity of the LTT are not fixed values and vary with the culprit agent, clinical context, and testing conditions. A negative result does not exclude the possibility of drug allergy, but a positive result should be considered supportive evidence. In our patient, SI of zinc exceeded the diagnostic threshold (>3), whereas SI of other agents remained below the diagnostic threshold. Although the LTT is increasingly used in the assessment of delayed drug hypersensitivity reactions, it has well-recognized limitations. According to the ENDA/EAACI Drug Allergy Interest Group position paper, LTT should be regarded as a complementary diagnostic tool, as its sensitivity and specificity vary depending on the implicated drug, clinical phenotype, and testing conditions. False-negative results may occur due to low frequencies of drug-specific memory T cells, ongoing or recent immunosuppressive treatment, or suboptimal timing of testing, particularly when performed shortly after severe cutaneous adverse reactions such as Stevens–Johnson syndrome and toxic epidermal necrolysis.13, 14 Conversely, false-positive results may arise from non-specific lymphocyte activation or hypersensitivity to formulation excipients, especially in non–β-lactam drugs and nutritional supplements, for which in vitro assay standardization remains limited.14 To minimize these limitations, testing was performed 8 weeks after recovery, using appropriate controls and predefined stimulation index criteria. The LTT was conducted with the same zinc preparation that had been administered to the patient. Although the LTT is not standardized for zinc, in this case it was interpreted as supportive evidence. Other potential triggers were excluded by detailed medication history, negative patch testing for alternative agents, exclusion of infectious causes, and ALDEN-based causality assessment. In conclusion, to our knowledge, this case is the first pediatric report of toxic epidermal necrolysis induced by oral zinc sulfate, with LTT supporting a T-cell-mediated mechanism. It highlights that even common supplements may rarely trigger severe cutaneous reactions, underscoring the need to consider recent supplement use in SJS/TEN evaluation. Kezban İpek Demir: Conceptualization; investigation; funding acquisition; writing – original draft. Burcu Özkan Kırgın: Methodology; validation. Zeynep Güleç Köksal: Writing – original draft; writing – review and editing. Fatma Nur Öz: Software; data curation. Sultan Göncü: Software; formal analysis. Büşra Daştan İnce: Visualization; formal analysis. Ayşegül Ertuğrul: Formal analysis; validation. Ezgi Ulusoy Severcan: Funding acquisition; project administration. Seda Şirin: Methodology; validation; writing – review and editing; resources; supervision. No specific funding was received for this work. All authors declare no conflicts of interest. As this is a single case report involving no experimental procedures, ethics committee approval was not required. Informed consent was obtained from parents. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/pai.70296.