SPONTANEOUS SOFT DRUSEN REGRESSION WITHOUT ATROPHY AND THE DRUSEN OOZE.

Objective Or Purpose: To determine the incidence of spontaneous soft drusen regression without atrophy (DRwoA) in patients with intermediate or atrophic age-related macular degeneration (iAMD, aAMD) and evaluate associated events and offer potential explanations.

Design: A retrospective review of the imaging of a consecutive series of 640 eyes from 320 patients with AMD who had at least 2 years of follow-up.

Subjects: 427 eyes from 262 patients with iAMD or aAMD and no present or past exudative AMD.

Methods: Retinal imaging included infrared reflectance imaging (IR), fundus autofluorescence (FAF), spectral-domain OCT (SD-OCT), and color fundus photography (CFP).

Main Outcome Measures: Drusen regression without atrophy (DRwoA) with integrity of the retinal pigment epithelium (RPE) and repositioning over Bruch's membrane. Additionally, drusen that would collapse with atrophy (DCwA) in the same area simultaneously were named "sentinel" DCwA. The reversibility of features of incomplete retinal pigment epithelium and outer retinal atrophy (iRORA), as well as areas ("halos") of DRwoA around the "sentinel" drusen.

Results: Among the 427 eyes, 53 events of DRwoA were found, representing 24,17% of the eyes with soft drusen. In 50 cases (94,33%), a "sentinel" DCwA in the vicinity was found. In 58% of the cases, a well-identifiable halo of drusen disappearance around the "sentinel" DCwA was well visible.

Conclusions: DRwoA is a frequently observed phenomenon linked to soft drusen and almost invariably occurs near a "sentinel" DCwA. The coalescence of the drusen and the spatial and temporal association of the DRwaA and the DCwA strongly suggest that the drusen material of DRwoA escapes to a contiguous "sentinel" DCwA. While the hypothesis of the disappearance of drusen material due to RPE death may explain the DCwA, it fails to account for DRwoA. Instead, the "drusen ooze" hypothesis, which posits the movement of drusen content to the subretinal space through RPE defects, may explain both the DCwA and the DRwoA. This hypothesis offers insights into the reversibility of iRORA features and suggests that therapies targeting drusen material removal before RPE disruptions could potentially prevent atrophy secondary to soft drusen collapse.