Retinal Remodeling in Retinopathy of Prematurity

Abstract

 

Purpose

 

The sensitivity of the electroretinographic (ERG) a-wave, which originates in photoreceptors, is low in mature eyes with a history of retinopathy of prematurity (ROP). However, the sensitivity of the ERG b-wave, which originates in postreceptor retina, often becomes normal with age (Harris et al., 2011, Doc Opthalmol 122(1):19). Furthermore, near the ‘rod ring,’ teenaged patients with a history of ROP have fairly normal retinal sensitivity but larger areas for complete scotopic spatial summation than do either preterm individuals with no history of ROP or term born controls (Tavormina et al., 2013, ARVO Abs. 5850). Collectively, these data suggest reorganization of the peripheral, postreceptor circuitry that trades spatial resolution for increased sensitivity. Thus, the retinal layers near the rod ring were studied by adaptive-optics optical coherence tomography (OCT) in this preliminary study.

 

Methods

 

Spectral-domain OCTs of a 1° slice located 18° nasal of the fovea were obtained in 12 term-born controls, 4 preterm (≤32 weeks gestation) subjects and 3 subjects with a history of ROP. Flattened, aligned and averaged frames were manually marked in ImageJ to delineate the following layers: 1) nerve fiber, 2) inner plexiform and ganglion cell, 3) inner nuclear, 4) outer plexiform, 5) outer nuclear, 6 and 7) respective mitochondria sparse and mitochondria dense portions of the inner segment, 8) outer segment, and 9) retinal pigment epithelium and choriocapillaris. Mean layer thickness was plotted for each group and inspected for patterns. The total thickness of the aggregate ‘postreceptor’ layers (1-4) and ‘photoreceptor’ layers (5-9) were tested for significant differences by two-factor (group, depth) repeated measures ANOVA, followed by Tukey’s HSD post-hoc test.

 

Results

 

Inspection of the means revealed that, relative to term, ROP layers 1-4 (postreceptor) were notably thicker, layers 5-8 (photoreceptor) were thinner, and 9 was almost unchanged. Correspondingly, there was no main effect of group (P=0.914) but there was a highly significant group×depth interaction (P<0.001) and the HSD test revealed significantly thicker postreceptor retina and thinner photoreceptor retina in ROP than in term subjects; in preterm subjects, neither depth was distinguishable from ROP or term.

 

Conclusions

 

There is anatomic evidence for postreceptor remodeling compensatory to loss of photoreceptor input in the peripheral ROP retina.

 

Keywords: 706 retinopathy of prematurity • 551 imaging/image analysis: non-clinical • 650 plasticity