Systemic Inflammatory Markers, Thyroid Autoimmunity, and Neuroretinal Structure in Graves' Disease: A Comparative OCT Study According to Ophthalmopathy Status.

Abstract

Background: Graves' disease is a systemic autoimmune disorder that may affect ocular structures even in the absence of clinically apparent ophthalmopathy. Optical coherence tomography (OCT) enables objective assessment of neuroretinal structures, but the relationships between OCT findings, systemic inflammatory indices, thyroid hormone levels, and thyroid autoimmunity markers in Graves' disease remain incompletely defined. This study aimed to evaluate the associations between OCT-derived neuroretinal measurements and systemic inflammatory, hormonal, and autoimmune markers according to ophthalmopathy status. Methods: This retrospective cross-sectional study included 161 participants: 60 healthy controls, 50 Graves' disease patients with ophthalmopathy (GO+), and 51 Graves' disease patients without ophthalmopathy (GO-). Central macular thickness (CMT) and circumpapillary retinal nerve fiber layer (cpRNFL) thickness measurements were obtained using spectral-domain OCT. Macular RNFL (mRNFL) thickness was not evaluated. Laboratory evaluation included complete blood count, glucose, urea, creatinine, aspartate aminotransferase (AST), alanine aminotransferase (ALT), total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides, C-reactive protein (CRP), thyroid-stimulating hormone (TSH), free triiodothyronine (fT3), free thyroxine (fT4), anti-thyroid peroxidase (anti-TPO), and anti-thyroglobulin (anti-Tg) antibodies. Results: In the primary three-group OCT analysis, significant differences were observed for superior (p = 0.010, q = 0.0125), inferior (p < 0.001, q = 0.0024), temporal (p = 0.002, q = 0.0050), and nasal cpRNFL thickness (p = 0.006, q = 0.0092), and all remained significant after FDR correction; CMT did not differ among the groups (p = 0.925, q = 0.925). The lowest cpRNFL thickness values were observed in the GO- subgroup, particularly in the inferior quadrant (116.39 ± 27.52 vs. 136.65 ± 20.01 µm, post hoc p < 0.001); however, this finding should be interpreted cautiously because standardized GO activity and severity staging was not consistently available. Systemic inflammatory indices were largely similar between groups, whereas CRP levels were higher in Graves' disease in the exploratory analysis (3.87 ± 3.40 vs. 1.86 ± 1.11 mg/L, nominal p < 0.001). Several weak correlations between cpRNFL parameters and fT4, CRP, and anti-TPO reached nominal significance; however, none remained statistically significant after FDR correction. In multivariable analysis restricted to patients with Graves' disease, GO+ status remained associated with greater inferior cpRNFL thickness compared with GO- status, whereas anti-TPO was not associated with inferior cpRNFL thickness after adjustment. Conclusion: Sectoral cpRNFL differences were observed among the control, GO+, and GO- groups, whereas CMT remained comparable. However, the lower cpRNFL measurements in patients classified as GO- should be interpreted cautiously because of retrospective ophthalmopathy classification, potential misclassification, differential exclusion rates, and residual confounding. These findings are strictly exploratory and hypothesis-generating and require prospective validation. Biomarker-OCT correlations did not remain statistically significant after FDR correction.