Typical DRESS patients would have been exposed to a high-risk medication . A retrospective analysis of 45 patients with DRESS showed that the most common causative drug group was antibiotics (28.9%), followed by anticonvulsants (26.7%) . Our patient was exposed to the antibiotic ornidazole, which is not a conventional antibiotic; however, several case reports do show that it is related to fixed drug eruptions . Two other nitro-imidazole drugs, tinidazole and metronidazole, have also been reported to be causes of DRESS [10, 11]. According to the scoring system for classifying DRESS , our patient had a score of 5, indicating that she probably suffered from DRESS. In another retrospective study, about 12% of patients with DRESS had not been exposed to a recognized high-risk drug .
In our case, the patient developed FT1D 2 weeks after DRESS flare-up. Her serum level of proinsulin and C-peptide were both undetectable soon after FT1D onset. A recent study revealed that in immune-mediated type 1 diabetes of long duration, the ability of the pancreas to secrete proinsulin persists even at an undetectable serum C-peptide level . Thus, beta islet destruction in FT1D is more severe than that in immune-mediated type 1 diabetes.
In a retrospective study which recruited 145 patients with DRESS, five patients subsequently developed FT1D . An increasing body of data have revealed that defective T regulatory (Treg) cells play a central role in FT1D [14, 15]. Supporting evidence for the role of Treg cells comes from one study which showed that the FT1D patients in the study had lymphocytic infiltration in both the endocrine and exocrine pancreas, with 98% of them manifesting elevated levels of serum pancreatic enzymes , as the case of our patient (amylase 132 U/L, lipase 1097 U/L).
Treg cell dysfunction signified the resolution stage of DRESS. Treg cells are vigorously activated and expanded in the acute stage of DRESS, while they are functionally defective in the resolution stage . This functional defectiveness may be responsible for the subsequent unopposed inflammatory insult to a specific organ, such as FT1D.
Other theories proposed for DRESS and FT1D include viral reactivation and HLA susceptibility . Yoneda et al. performed immunohistochemistry of the autopsied pancreas of a patient who developed FT1D 3 weeks after DRESS; they detected human cytomegalovirus (HCMV)-positive cells together with macrophages and CD4+/CD8+ T lymphocytes in islets and exocrine areas, indicating that HCMV infection was associated with destruction of the beta cells . As in our case the higher ratio of CD8+ DR+/CD8 and CD8+ CD38+/CD8+ suggested viral infection (Table 2). Onuma et al reviewed 15 cases of DRESS complicated with FT1D reported in the literature and found a high frequency of HLA B62 in those patients . However, our patient did not have the HLA B62 subtype (ESM Table 2). Thus, more data are needed to study the relationship between HLA subtype and DRESS complicated with FT1D.
The low free triiodothyronine/free thyroxine (FT3/FT4) ratio and thyroid iodine absorption rate suggested inflammatory damage to the thyroid and its destruction induced by iodine overload, possibly due to leakage of thyroxine out of the destroyed follicular cells of the thyroid; this mechanism would be consistent with that of IIT . The patient’s anti-TPO and anti-Tg antibodies were positive, indicating that she had Hashimoto’s disease, which is one of the risk factors for IIT. More importantly, DRESS itself disrupted the immune system and rendered the patient vulnerable to IIT, which further elucidates that immune dysregulation plays a central role in both IIT and FT1D.
The case described herein is the first in which IIT is reported as a sequela of DRESS. Physicians should be aware that as IIT could be a sequela of DRESS, iodinated substances are inappropriate for the treatment of these patients.