Article Text
Abstract
Background/aims Handheld swept-source optical coherence tomography (SS-OCT) was previously used to measure foveal maturity through semiautomated methods in awake premature infants. This study assesses the relationship between foveal maturity and retinopathy of prematurity (ROP) severity.
Methods This is a prospective, observational study using handheld SS-OCT imaging of premature infants undergoing ROP screening. A semiautomated segmentation algorithm measured foveal angle, total retinal cross-sectional area, and retinal and choroid thicknesses at the fovea and parafovea, excluding significant macular oedema. Measures were correlated with ROP severity using a mixed model approach.
Results 99 eye imaging sessions from 26 preterm infants were included. Average birth weight was 1057.6±324.8 g. Gestational age and postmenstrual age at imaging were 29.1±3.0 and 36.1±3.8 weeks, respectively. Stage 3 ROP occurred in 16/99 (16.2%) exams. Shallow foveal angle and higher inner retinal fovea/parafoveal (F/P) ratio correlated with more posterior ROP zone (zone 2: 119.2°±20.9° vs zone 3: 96.5°±16.7° and zone 2: 0.6±0.2 vs zone 3: 0.4±0.1, respectively, both p<0.001). Inner retinal F/P ratio was greater in eyes with higher ROP stage (stage 3: 0.6±0.2 vs stage 0 or mature: 0.4±0.2, p=0.03) but this relationship was not seen for outer retina (p=0.29). Larger retinal cross-sectional area coincided with worse ROP stage (stage 3: 1.9±0.06 mm2 vs stage 0 or mature: 1.6±0.04 mm2, p<0.001) and zone (zone 2: 1.6±0.04 mm2 vs zone 3: 1.7±0.04 mm2, p=0.01).
Conclusions SS-OCT semiautomated measurements of inner retinal F/P ratio, foveal angle and retinal cross-sectional area may serve as ROP severity biomarkers.
- Retina
- Imaging
Data availability statement
Data are available on reasonable request.
Statistics from Altmetric.com
Footnotes
Contributors EKT, KEL and MTC collected the semiautomated OCT segmentation data. EKT, KEL and MTC analysed and interpreted the data and drafted and provided critical revisions to the manuscript. SL and TRM collected the OCT image data. ATL, JPK, KT-H and MTC collected the ROP data. TL, YZ and RW designed the software for image processing. RW designed the hand-held OCT. LD provided statistical analysis. MTC designed the study and is the guarantor of the study. All authors read and approved the manuscript.
Funding Alcon Research Institute Young Investigator Award, unrestricted grants from Research to Prevent Blindness and National Institutes of Health (EY00130) to the University of Washington Department of Ophthalmology, Latham Vision Research Innovation Award, Violet Sees.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.