Tattoo Granuloma Associated With Uveitis: A Useful Temporary Entity Pending Further Clarification—Response

We thank Dr. Kluger for his response and for the opportunity to reply [1]. Our comment that no definition of systemic sarcoidosis was used referred to his 2018 review [2], where the definition from his 2013 paper (granulomatous involvement of two organs) was not included [3]. Our two patients with hilar lymphadenopathy had no lung involvement, no chest symptoms and no tissue confirmation of granulomata [4]. Hilar adenopathy alone, without granulomas, does not meet Kluger's criteria for systemic sarcoidosis [3]. Ophthalmologists currently use the SUN classification of 2021 which requires a typical ophthalmic sarcoid presentation plus biopsy-confirmed non-caseating granulomata or bilateral hilar adenopathy. By this definition, all TAU cases would likely constitute sarcoidosis if the skin were biopsied. We did not use the TAGU label for clinical reasons [4]. Early presentations had non-granulomatous uveitis and most tattoos were not biopsied. TAU is a more useful working label because it covers all presentations regardless of histology, does not assume a mechanism, and reflects how the diagnosis is made in practice. Hao et al. have recently shown in six TAU patients that active TAU and sarcoidosis have distinct molecular profiles [5]. CXCR5 downregulation was specific to active TAU and absent in sarcoidosis, and the clonal CD8+ T cell expansion seen in sarcoidosis was not found in TAU [5]. Both conditions shared elevated TNF expression in monocytes and B cell enrichment of genes related to TNF signalling which may explain why both respond to anti-TNF treatment. In contrast, enriched TNF signalling via the NF-κB pathway in CD4+ T cells was present in sarcoidosis but not TAU [5]. This molecular evidence suggests TAU is immunologically distinct from sarcoidosis rather than simply a limited form of it. Until molecular or immunologic techniques allow definitive classification, lumping rather than splitting these diagnoses remains most clinically useful. Our series, like Kluger's reviews, was retrospective [2, 4]. TAU was often not recognised at first presentation and testing was commonly started by non-uveitis specialists. When a uveitis specialist made the diagnosis early, fewer tests were done. Serum calcium's low sensitivity and specificity in uveitis and granulomatous disease, with elevation in under 10% of patients with systemic sarcoid limits its utility in ophthalmology. Our paper suggested a baseline CT chest, serum ACE and ECG for future cases [4]. Treating the uveitis systemically would likely also treat any asymptomatic sarcoidosis, so investigating fully is currently an academic rather than clinically relevant pursuit, which may not represent cost effective use of health resources. While it is commonly asserted that 90% of sarcoidosis patients have chest findings, our data suggest this may not apply when the disease trigger is through the skin [4]. Active TAU is associated with downregulation of CXCR5 on B cells and CXCR3 on T cells, changes that may limit immune cell trafficking to lymph nodes and constrain disease to tissue-specific rather than systemic pattern [5]. Prospective CT chest at presentation combined with biopsy of the inflamed tattoo and immunologic blood analysis may help clarify this. On Table 2, we accept that comparing Dr. Kluger's group, drawn from published case reports [2], with our full retrospective cohort is not straightforward [4]. The table was intended to show both the phenotypic overlap between the two groups and the more ophthalmic nature of disease in our cohort. Treatment comparison was not meaningful given the evolution of uveitis management since 2018. We agree with Dr. Kluger that the relationship between TAU and sarcoidosis warrants further study [1]. We are planning prospective data collection including ink type, tattoo history and timing of skin and eye disease, and are collaborating with groups studying the proteomics of, and immune mechanisms behind tattoo-associated inflammation. The authors have nothing to report. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.