The aim of this observational study is to evaluate retinal microvascular alterations in patients with thyroid-associated orbitopathy (TAO) using optical coherence tomography angiography (OCTA). The specific objectives are:

• To assess the impact of extraocular muscle (EOM) enlargement on vessel density in the superficial and deep retinal capillary plexuses in patients with TAO.
• To determine the effect of increased orbital fat volume on superficial and deep retinal vessel density in TAO.
• To evaluate the relationship between disease activity, as measured by the Clinical Activity Score (CAS), and retinal microvascular density in patients with TAO.

After obtaining written informed consent and confirming eligibility, all patients with TAO and controls underwent a detailed medical history review and ophthalmologic examination. Collected clinical data included age and sex. For patients with TAO collected data also oncluded duration of Graves' hyperthyroidism (GH) and orbitopathy (GO), as well as serum levels of thyroid-stimulating hormone (TSH), triiodothyronine (T3), thyroxine (T4), thyrotropin receptor antibodies (TRAb), and anti-thyroid peroxidase antibodies (anti-TPO). For all participants a comprehensive ophthalmologic assessment was performed, which included evaluation of the periocular region, ocular motility, pupillary reflexes, best-corrected visual acuity (BCVA), slit-lamp biomicroscopy, intraocular pressure (IOP) measurement using Goldmann applanation tonometry in primary and upward gaze, exophthalmometry with a Hertel device, and dilated fundus examination.

Disease activity in patientss with TAO was assessed using the Clinical Activity Score (CAS), following the European Group on Graves' Orbitopathy (EUGOGO) guidelines.

MRI images of orbit in TAO patients acquired using 1.5T and 3T scanners (Siemens Magnetom Avanto Fit 124 and Magnetom Skyra, Siemens Healthcare, Germany) during TAO evaluation were reviewed manually by an experienced radiologist using the PACS imaging software (Sectra 126 IDS7, Version 24.1, Linköping, Sweden). Measurements were performed on T1- and T2-weighted images in both coronal and axial planes.The maximum diameter of each of the four rectus muscles (medial, lateral, superior, and inferior) was measured at its thickest midpoint. Horizontal diameters of the medial and lateral rectus muscles were obtained on axial images, while vertical diameters of the superior and inferior rectus muscles were measured on coronal sections.

Exophthalmos was determined as the perpendicular distance between the anterior surface of the cornea and the line connecting both lateral orbital rims on the slice where the globe and optic nerve were most clearly visualized. Orbital fat thickness was measured as the maximum distance from the lateral border of the medial rectus muscle to the medial orbital wall. Based on radiological characteristics, patients with TAO were classified into three subtypes: fat-predominant, muscle-predominant, and mixed type.

Optical coherence tomography angiography (OCTA) was performed in all paticipants (patients with TAO and controls) using the OptoVue XR Avanti device equipped with the split-spectrum amplitude-decorrelation angiography (SSADA) algorithm (software version 2014.2.0.90), with the Angio Retina 6×6 mm protocol. Retinal vessel density was analyzed at two vascular levels-the superficial capillary plexus and the deep capillary plexus. Following the instrument's standardized protocol, the macula was divided into nine regions: a central foveal zone, parafoveal ring, and perifoveal ring, each subdivided into superior, inferior, nasal, and temporal quadrants. The foveal avascular zone was also measured.

Inclusion Criteria:

• adult patients (≥18 years) of both sexes with a confirmed diagnosis of TAO treated at University Hospital Osijek.

Exclusion Criteria:

• anterior segment pathology (e.g., central keratopathy),
• primary open- or closed-angle glaucoma,
• posterior segment disorders (e.g., diabetic retinopathy, age-related macular degeneration, retinal vascular occlusion),
• previous ocular trauma or uveitis,
• refractive errors exceeding ±6.0 Diopter sphere or ±3.0 Diopter cylinder
• media opacities that compromised angio optical OCTA image quality,
• contraindication for orbital MRI

Patients with TED followed at Department of Ophthalmology, University Hospital Osijek, Department of Ophthalmology, Croatia.