Choroidal Thickness in Thyroid Eye Disease: Comparison With Controls and Application in Diagnosing Non-Inflammatory Active Disease.

Introduction Choroidal thickness is known to vary in various systemic diseases. In the current study, we aim to report the differences in choroidal thickness in thyroid eye disease (TED) and normals and its discriminatory value for differentiating various stages of TED. Methods Prospective, cross-sectional, non-interventional imaging study. In an institutional practice, 102 eyes of 51 patients were included and divided into five groups: normal controls (C), inactive TED (I), active TED (A), non-inflammatory active TED (NIA) and systemic thyroid disorder but no TED (SYS). Choroidal images were acquired using the swept-source optical coherence tomography (Topcon DRI OCT Triton) with automatic layer segmentation which provided an automatic measurement of the subfoveal choroidal thickness and the mean in nine subfields based on the Early Treatment Diabetic Retinopathy Study (ETDRS) grid. One-way analysis of variance (ANOVA), Youden index and area under the receiver operating characteristic curves (AUROC) were reported. Results Central choroidal thickness in the A group was 279±37.52 microns and in the NIA group was 302.5±59.22 microns. Both were comparable to each other and significantly higher than the C, I and SYS groups (p<0.001). All ETDRS sub-fields showed significant AUROC to distinguish NIA from I. Most significant Youden index was for the inner nasal and central ETDRS subfields (0.55 and 0.61 respectively). Inner nasal sub-field showed 100% specificity while the central sub-field, showed 86.5% for predicting NIA. At a choroidal thickness of >266 microns, the central sub-field had the strongest discriminatory potential to predict NIA.  Conclusion Choroidal thickness is greater in active and non-inflammatory active TED. The inner nasal and central ETDRS sub-fields have value in differentiating the non-inflammatory active TED eyes from the inactive eyes.