Abstract
Purpose: :
The hallmark of ROP is abnormal retinal blood vessels, but neural dysfunction in receptor and post–receptor retina is also a feature ROP. If, and how, the vascular and neural abnormalities are related is unknown. We sought to evaluate longitudinal measures of the retinal vasculature and neural retinal function in two rat models of ROP for signifiant neurovascular relationships.
Methods: :
The first model was induced by exposing rats (ages 7–14 days) to continuous 75% oxygen. The second was induced by exposing rats (ages 0–14 days) to alternating 50/10% oxygen. Room–air raised controls were also studied. Function of the neural retina was derived from serial recordings of ERG responses in dark adapted rats at age 20, 30, and 60 days; rod (S, RmP3) and post–receptor (log σ, Vmax) response parameters were calculated. In the same animals in the same sessions, digital fundus photographs were obtained, and the tortuosity index (TI, ratio of actual to straight line length) of the retinal vessels was calculated. Within subjects ANOVA compared group and age for each parameter. Regression analyses evaluated functional and vascular parameters for significant relationships.
Results: :
In the 50/10% model TI was initially high, then decreased in every rat to nearly normal values by 60 days. In the 75% model TI remained high and did not significantly vary with age. In both ROP models, significant deficits in saturated rod response amplitude, RmP3, were similar and stable. Deficits in RmP3 and Vmax, the post–receptor response amplitude, were correlated. In the 20 to 60 day age range, although rod sensitivity, S, increased modestly in the 50/10% model and in controls, deficits persisted in the 75% model. The post–receptor sensitivity parameter, log σ, increased significantly in both ROP models. Analysis of all subjects at initial assessment (20 days) showed TI and the sensitivity parameters of the neural retina (S , log σ) were correlated (Spearman), but TI and amplitude parameters (RmP3, Vmax) were not. With increasing age, improvements in log σ were correlated with the decreases in TI observed in the 50/10% model. In the 75% model, the increase in log σ was independent of TI.
Conclusions: :
The improvements in the retinal vessels (TI) and photoreceptor sensitivity in the 50/10% model, and the persistence of blood vessel abnormalities and photoreceptor sensitivity in the 75% model, go hand in hand, illustrating a consistent neurovascular relationship in the ROP disease process. In contrast, post–receptor response parameters do not have a consistent relation to TI.
Keywords: retinopathy of prematurity • electroretinography: non-clinical • neovascularization