Possible retinotoxicity of long-term vardenafil treatment

Phosphodiesterase (PDE) inhibitors - such as vardenafil - are used primarily for treating erectile dysfunction via increasing cyclic guanosine monophosphate (cGMP) levels. Recent studies have also demonstrated their significant cardioprotective effects in several diseases, including diabetes, upon long-term, continuous application. However, PDE inhibitors are not specific for PDE5 and also inhibit the retinal isoform. A sustained rise in cGMP in photoreceptors is known to be toxic; therefore, we hypothesized that long-term vardenafil treatment might result in retinotoxicity. The hypothesis was tested in a clinically relevant animal model of type 2 diabetes mellitus. Histological experiments were performed on lean and diabetic Zucker Diabetic Fatty rats. Half of the animals were treated with vardenafil for six months, and the retinal effects were evaluated. Vardenafil treatment alleviated rod outer segment degeneration but decreased rod numbers in some positions and induced changes in the interphotoreceptor matrix, even in control animals. Vardenafil treatment decreased total retinal thickness in the control and diabetic groups and reduced the number of nuclei in the outer nuclear layer. Müller cell activation was detectable even in the vardenafil-treated control animals, and vardenafil did not improve gliosis in the diabetic group. Vardenafil-treated animals showed complex retinal alterations with improvements in some parameters while deterioration in others. Our results point towards the retinotoxicity of vardenafil, even without diabetes, which raises doubts about the retinal safety of long-term continuous vardenafil administration. This effect needs to be considered when approving PDE inhibitors for alternative indications.

The Predictive Role of Thyroid Hormone Levels for Early Diabetic Retinal Changes in Experimental Rat and Human Diabetes

Purpose

In diabetic subjects, early visual functional alterations such as color vision deficiencies (CVDs) are known to precede clinically apparent diabetic retinopathy. Prominent photoreceptor outer segment degeneration and an increase in the number of retinal dual cones (co-expressing S- and M-opsins simultaneously) have been described in diabetic rat models, suggesting a connection with the development of CVDs. As cone opsin expression is controlled by thyroid hormones, we investigated the diabetic retina in association with thyroid hormone alterations.

Methods

In rat models of type 1 and 2 diabetes, dual cones were labeled by immunohistochemistry, and their numbers were analyzed in relation to free triiodothyronine (fT3) and free thyroxine (fT4) levels. Quantification of dual cones was also performed in human postmortem retinas. Additionally, a cross-sectional case–control study was performed where thyroid hormone levels were measured and color vision was assessed with Lanthony desaturated D15 discs.

Results

A higher number of dual cones was detectable in diabetic rats, correlating with fT4 levels. Dual cones were also present in postmortem human retinas, with higher numbers in the three diabetic retinas. As expected, age was strongly associated with CVDs in human patients, and the presence of diabetes also increased the risk. However, the current study failed to detect any effect of thyroid hormones on the development of CVDs.

Conclusions

Our results point toward the involvement of thyroid homeostasis in the opsin expression changes in diabetic rats and human samples. The evaluation of the possible clinical consequences warrants further research.

Histological Evaluation of Diabetic Neurodegeneration in the Retina of Zucker Diabetic Fatty (ZDF) Rats

In diabetes, retinal dysfunctions exist prior to clinically detectable vasculopathy, however the pathology behind these functional deficits is still not fully established. Previously, our group published a detailed study on the retinal histopathology of type 1 diabetic (T1D) rat model, where specific alterations were detected. Although the majority of human diabetic patients have type 2 diabetes (T2D), similar studies on T2D models are practically absent. To fill this gap, we examined Zucker Diabetic Fatty (ZDF) rats - a model for T2D - by immunohistochemistry at the age of 32 weeks. Glial reactivity was observed in all diabetic specimens, accompanied by an increase in the number of microglia cells. Prominent outer segment degeneration was detectable with changes in cone opsin expression pattern, without a decrease in the number of labelled elements. The immunoreactivity of AII amacrine cells was markedly decreased and changes were detectable in the number and staining of some other amacrine cell subtypes, while most other cells examined did not show any major alterations. Overall, the retinal histology of ZDF rats shows a surprising similarity to T1D rats indicating that despite the different evolution of the disease, the neuroretinal cells affected are the same in both subtypes of diabetes.

Study of retinal neurodegeneration and maculopathy in diabetic Meriones shawi: A particular animal model with human-like macula

The purpose of this work was to evaluate a potentially useful animal model, Meriones shawi (M.sh)-developing metabolic X syndrome, diabetes and possessing a visual streak similar to human macula-in the study of diabetic retinopathy and diabetic macular edema (DME). Type 2 diabetes (T2D) was induced by high fat diet administration in M.sh. Body weights, blood glucose levels were monitored throughout the study. Diabetic retinal histopathology was evaluated 3 and 7 months after diabetes induction. Retinal thickness was measured, retinal cell types were labeled by immunohistochemistry and the number of stained elements were quantified. Apoptosis was determined with TUNEL assay. T2D induced progressive changes in retinal histology. A significant decrease of retinal thickness and glial reactivity was observed without an increase in apoptosis rate. Photoreceptor outer segment degeneration was evident, with a significant decrease in the number of all cones and M-cone subtype, but-surprisingly-an increase in S-cones. Damage of the pigment epithelium was also confirmed. A decrease in the number and labeling intensity of parvalbumin- and calretinin-positive amacrine cells and a loss of ganglion cells was detected. Other cell types showed no evident alterations. No DME-like condition was noticed even after 7 months. M.sh could be a useful model to study the evolution of diabetic retinal pathology and to identify the role of hypertension and dyslipidemia in the development of the reported alterations. Longer follow up would be needed to evaluate the potential use of the visual streak in modeling human macular diseases.