Microaneurysm Counting as a Biomarker for the Hyperperfusion Stage of Nonproliferative Diabetic Retinopathy

This study aimed to investigate the utility of microaneurysm (MA) counting as a tool for characterizing the hyperperfusion stage of nonproliferative diabetic retinopathy (NPDR) and to examine the hypothesis that MAs can serve as a surrogate biomarker for the presence of intraretinal microvascular abnormalities (IRMAs).

Forty-nine (n = 49) eyes with type 2 diabetes mellitus with NPDR were included in this analysis: 12 with Early Treatment Diabetic Retinopathy Study (ETDRS) levels 43 and 37 with levels 47-53. Automated MA detection was performed using the RetmarkerDR software (Retmarker SA, Meteda Group, Italy), alongside manual detection, both done in the central retina (field 2). Based on MA counts, microaneurysm turnover (MAT) was computed. IRMAs were manually counted based on swept-source optical coherence tomography (SS-OCT) angiography on PLEX® Elite 9000 (ZEISS, Dublin, CA, USA). The statistically significant differences between ETDRS groups were studied by comparing Mann-Whitney U test p values (significance value < 0.05). The correlation between the presence of MAs and IRMAs and MAT and IRMAs was examined using Spearman correlation analysis.

There was an observed increase in the number of IRMAs, MAs, and MAT values as NPDR progressed, independently of the counting method used. Specifically, this increase was noted when transitioning from ETDRS groups characterized by the predominance of the hypoperfusion stage (ETDRS 43) to those associated with the hyperperfusion stage (ETDRS 47-53). When MAs were counted manually, a moderate correlation was identified between the number of MAs and the presence of IRMAs (ρ = 0.40; p value = 0.005). Additionally, a similar correlation was found between MAT and the presence of IRMAs (ρ = 0.43; p value = 0.002).

This study underscores the potential relevance of MAs as a pivotal indicator of the hyperperfusion stage of NPDR and supports their role as surrogate biomarkers for IRMAs. These results suggest a role for MA counting in the assessment and management of diabetic eye disease.