ObjectiveTo explore the feasibility to establish a novel retinal ischemia reperfusion (RIR) injury model by applying pars plana vitrectomy (PPV) combined with retinal vascular counterpulsation in the central retinal artery occlusion (CRAO) model of New Zealand rabbit.

MethodsTwenty New Zealand adult rabbits were randomly divided into two groups by random number table, 10 for each group.Rabbits in the laser group were treated with retinal laser photocoagulation only, while rabbits in the counterpulsation group were treated with PPV combined counterpulsation.The right eye of each New Zealand rabbit was used as the experimental eye and the left eye was used as the normal control (the normal control group). Fundus fluorescence angiography (FFA) , oxygen partial pressure (PO ₂) in vitreous cavity was performed to assess the recovery status of perfusion.Scotopic 3.0 oscillatory potentials (OPs) in electroretinogram (ERG) were used to evaluate the retinal function, and retinal pathological sections were used to evaluate the structural changes in the retina.The use and care of the animals complied with the Statement of the Association for Research in Vision and Ophthalmology (ARVO), and the study was approved by the Animal Research Committee of Zhongshan Ophthalmic Center, Sun Yat-sen University.

ResultsIn the counterpulsation group, retinal reperfusion was observed during counterpulsation processure.FFA examination at 2 hours after counterpulsation showed reperfusion of retinal blood stream in all the eyes.Retinal artery filling, followed by venous filling was observed in the early stage, with no delay in filling and no interruption of blood flow.The percentage of vitreous PO ₂ was significantly different among the counterpulsation group, the laser group and the normal control group at different time points ( F _group=330.87, P<0.001; F _time=985.70, P<0.001). The percentages of vitreous PO ₂ in the counterpulsation group at different time points was (18.67±6.29)%, (38.82±1.48)%, (57.33±4.25)%, (84.51±3.91)% and (89.20±2.97)%, which were significantly higher than that in the laser group ([23.24±1.95]%, [31.44±3.29]%, [40.21±3.05]%, [43.65±3.82]% and [58.07±2.93]%) (all at P<0.05). The percentage of OPs was significantly different among the counterpulsation group, the laser group and the normal control group at different time points ( F _group=164.09, P<0.001; F _time=447.91, P<0.001). The percentages of OPs of counterpulsation group at different time points after laser and after PPV were (47.23±2.73)%, (70.79±3.09)%, (78.39±3.63)%, (76.69±4.08)% and (82.18±1.78)%, which were significanthy higher than that in the laser group ([46.83±2.89]%, [55.32±1.58]%, [51.08±4.02]%, [52.32±6.59]% and [53.46±6.46]%) ( all at P<0.05). There was a lesser damage in inner retinal structure in pathological section in the counterpulsation group.The myelinated nerve fiber layer (MFL) was loose and a mass of vacuolar changes were observed in MFL.The structure of MFL, inner plexiform layer, inner and outer plexiform layer in the laser group were disordered, and the Müller cell nerve fibers were destroyed in the laser group.

ConclusionsA novel type of RIR injury model can be established by applying PPV combined with counterpulsation in the CRAO model of New Zealand rabbit.