Retinal Microvascular and Peripapillary Profiles in Mild Autonomous Cortisol Secretion Versus Nonfunctioning Adrenal Adenomas

Abstract

Background/Objectives: Mild autonomous cortisol secretion (MACS) is associated with increased cardiometabolic morbidity, but its relationship with the retinal microcirculation is unknown. We compared retinal and peripapillary microvascular parameters between MACS and nonfunctioning adrenal adenoma (NFA) and, secondarily, examined their associations with continuously modeled post-dexamethasone cortisol. Methods: In this single-center, retrospective, cross-sectional study, 98 patients (45 NFA, 53 MACS) were included, and optical coherence tomography angiography (OCTA) and structural OCT data from one eye per patient were analyzed. The primary outcome was whole-image deep capillary plexus (DCP) vessel density. Multivariable models were adjusted for age, sex, body mass index, and cardiometabolic comorbidities; secondary analyses were corrected using the Benjamini–Hochberg false discovery rate (q values). Results: OCTA was evaluable in 89 patients. The primary outcome did not differ significantly between groups in the main analysis (NFA vs. MACS, 48.64 ± 5.71% vs. 47.31 ± 5.77%; adjusted difference −1.29 percentage points, 95% CI −3.97 to 1.38; p = 0.339) or in any sensitivity analysis. In secondary analyses, superior retinal nerve fiber layer (RNFL) thickness was greater in MACS (β = 17.38 µm; q < 0.001) and increased with ln(post-dexamethasone cortisol) (β = 13.74 µm; q = 0.010). Higher ln(post-dexamethasone cortisol) was also associated with lower parafoveal DCP (β = −2.72 percentage points; p = 0.013) and inside-disc radial peripapillary capillary vessel density (β = −3.13 percentage points; p = 0.026), which did not survive multiplicity correction; no superficial capillary plexus sector was associated with either exposure (all p ≥ 0.37). Conclusions: MACS was not significantly associated with whole-image DCP vessel density. The secondary associations with continuous cortisol are cross-sectional and hypothesis-generating and require prospective confirmation.