Fig. 1: Poly(CEP) prevents scarring and retinal detachment in an experimental rabbit model of PVR.

a Representative ophthalmic images of rabbit posterior eye segments from non-operated (n = 5), poly(CEP)- (n = 8) and SF₆-filled (n = 5) eyes at day 14 post-surgery, and air-filled eye (n = 5) (experimental PVR model) at day 10 post-surgery. Colour fundus panel: Both non-operated and poly(CEP)-filled eyes showed a normal clear fundus with attached retina. SF₆-injected eye showed a hazy fundus with blurred view of vessels. Air-filled eye showed a completely detached retina (white arrow). IR panel: Both non-operated and poly(CEP)-filled eye displayed normal posterior segment. SF₆-injected eye showed fibrotic scar membranes indicated by the streaks of dark shadows (white arrowheads). Air-filled eye showed a dark shadow (white arrow) that was caused by the detached retina. Scale bar, 2 mm. SD-OCT (overview) panel: All groups except air-filled showed attached retina. Scale bar, 200 µm. SD-OCT (Enlarged from dotted box) panel: Both non-operated and poly(CEP)-filled eyes displayed normal stratified retinal layers. SF₆-injected eye showed presence of scar membrane (white arrowhead). Air-filled eye showed a completely detached retina. Scale bar, 50 µm. b Histopathological characterisation of enucleated rabbit retinas from all four groups at time points mentioned in (a). Retina from non-operated (n = 3) and poly(CEP)-filled eye (n = 4) looked morphologically similar with the presence of distinct and organised retinal layers and a continuous monolayered RPE (black arrowheads). SF₆-injected eye (n = 2) showed ERM causing retinal traction. Air-filled eye (n = 2) showed a folded retina with disorganised retinal layers and a scar membrane (black dotted outline). The native RPE was multi-layered. c Histology of retinas at 2 months post-surgery showed that poly(CEP)-filled eye (n = 4) maintained retinal layer organisation seen in the non-operated eye. Retina of SF₆-filled eye (n = 3) had presence of ERM and SRM (black dotted outlines). For both b, c artefactual tissue separation was observed due to sample processing (black stars). Scale bar, 100 µm. d Retinal cross sections from all groups at 2 weeks post-surgery were immunostained with fibrosis markers, α-SMA, and COL1A1. Non-operated and poly(CEP)-filled eyes (n = 4) showed minimal staining of all markers throughout all retinal layers. SF₆-filled eyes (n = 2) showed intense staining of both α-SMA and COL1A1 in both the pre-retinal fibrotic membrane (white arrow). Air-filled eyes (n = 2) had significant intra-retinal upregulation of COL1A1 with co-localisation of α-SMA at the epiretinal and outer retinal layers (white arrowheads). e Retinal sections from poly(CEP)- (n = 4) and SF₆-filled (n = 3) eyes at 2 months post-surgery were immunostained with α-SMA and COL1A1. Both poly(CEP)-filled and non-operated eyes (n = 3) had minimal α-SMA and COL1A1 staining, unlike SF₆-filled eyes which had positive staining both at the epiretinal (top retina panel) and subretinal zones (bottom RPE-CC panel). Nuclei was stained in blue with Hoechst stain. Scale bar for d, e, 50 µm. CC = Choriocapillaris. ERM = epiretinal membrane. GCL = ganglion cell layer. INL = inner nuclear layer. IPL = inner plexiform layer. IS/OS = photoreceptor inner/outer segments. ONL = outer nuclear layer. OPL = outer plexiform layer. PVR = proliferative vitreoretinopathy. RPE = retinal pigment epithelium. SRM = subretinal membrane.