Ocular fluorophotometry in human subjects and in swine - With particular reference to long-term pharmacokinetics

Abstract

The eye is surrounded by barriers controlling exchange of substance between eye and blood. These barriers have been reported affected in a number of disorders, where clinical manifestations and breakdown of the barriers has been observed simultaneously. The breakdown of the ocular barriers in diabetic retinopathy have been intensively explored due to increased risk for visual deterioration or blindness in these patients, but also due to the pronounced number of individuals suffering from this disorder. During the last 25 years, ocular barriers have mainly been studied by vitreous fluorophotometry, which include intravenous injection of fluorescein followed by measurement of this substance in the vitreous or the anterior chamber. From the vitreous fluorescein concentration profile and plasma fluorescein concentrations, the retinal penetration (permeability) for fluorescein can be obtained. These calculations take their offset in membrane physiological models and are based on a number of unverified assumptions. Due to these unverified assumptions and different mathematical approaches used in the anterior and posterior parts of the eye, it might be beneficial to obtain new models, which takes these circumstances into consideration. The purpose of the present study was to develop and apply a new (pharmacokinetic) model in normal subjects and in diabetic subjetcs with varying degree of retinopathy. This might improve our understanding of barrier affection in diabetic retinopathy and also increase our knowledge of specific treatment (photocoagulation and vitreous surgery). Free plasma fluorescein could be described by a biexponential decay curve, which is representative for an open two-compartment model. Free plasma fluorescence-glucuronide could be described with another biexponential function with a monoexponential decay, as long as detectable concentrations of the substance were present. The initial phase of disposition for fluorescein was due to distribution of fluorescein from the apparent central to the apparent peripheral compartment, in conjunction with elimination from the central compartment. The terminale phase of disposition was due to redistribution and elimination of fluorescein. The majority of fluorescein was located to the apparent peripheral tissue compartment of and only a minor fraction of fluorescein was located in the apparent central compartment. Vitreous fluorescein concentration time course could be described by a biexponential function in all normal and diabetic subjects. In the normal subjects, the maximal vitreous fluorescein concentration was found after 2-5 hr and followed by a monoexponential decay within the entire 24 hr examination period. From areas under vitreous and plasma fluorescein time/concentration curves a permeability index was defined. This index was found low in the group of normal subjects, while it increased significantly with increasing degree of retinopathy as reported in previous studies. The penetration rate constant (Kin) was found to decrease significantly with increasing degree of retinopathy. The penetration rate constant (Kin) was also found increased compared to the exchange velocity constant (K12) from the apparent central to the apparent peripheral compartment in normal subjects. This difference between Kin and K12 was not observed with increasing degree of retinopathy. In addition, the maximal vitreous fluorescein concentration in normal subjects was found less than the time corresponding fluorescein concentration in plasma. However, in diabetic subjetcs with pronounced background retinopathy or proliferative retinopathy, the maximal vitreous fluorescein concentration was found larger than the time corresponding plasma fluorescein concentration. These results might as a whole suggest the presence of an active transport system for fluorescein located within the retina. The presence of this active transport system might be an essential part of sustaining the tight barrier surrounding the posterior part of the eye in normal subjects. Its breakdown might be an essential part of the registered barrier breakdown in diabetes. The elimination rate constant of vitreous fluorescein (Kout) in normal and diabetic subjects was less than the elimination rate constant of fluorescein from the apparent peripheral tissue compartment. This is in accordance with ocular restriction of fluorescein elimination, possibly due to slow vitreous diffusion. The clinical importance of the defined permeability index was explored in 21 diabetic subjects, which were followed 5 years for proliferative development. Based on permeability indices from 26 normal subjects aged between 20 and 76 years the diabetic subjects were divided into two groups. One group included diabetic subjects with an increased permeability index (permeability) index above 95% confidence intervals of the normal group), while the other group included diabetic subjects with a normal permeability index (permeability index within or below the 95% confidence interval of the normal group). Following 5-year observation, 6 to 11 diabetic subjects with increased permeability index revealed proliferative development. This was in contrast to the group of diabetic subjects with a normal permeability index, where 0 of 8 developed proliferative retinopathy. Deterioration in visual acuity was not observed in these two groups within the 5 years followed-up period. An increased permeability index therefore points towards an increased risk for proliferative development, while there is no statistical correlation to visual acuity deterioration. The effect of panretinal photocoalgulation in proliferative diabetic retinopathy was explored by the present pharmaco-kinetic fluorophotometric method. Panretinal photocoalgulation was found to result in a delayed but increased fluorescein admittance to the eye. This has been explained by increased diffusibility of fluorescein from the choroid to the vitreous. It has been suggested that the documented beneficial effect of panretinal photocoagulation could be a result of increased accessibility for essential substance including oxygen. In addition, it was found that the barrier breaking-down in proliferative retinopathy was even more pronounced following panretinal photocoalgulation. It has therefore been suggested that the effect of treatment might be due to altered drug exchange across the barrier. The influence of possible active transport systems seems to fall, and the influence of possible diffusibility processes seems to increase. The influence of vitreous surgery on drug exchange to the eye was explored by the pharmacokinetic fluorophotometric method, applied on an experimental swine model. The blood-retinal barrier was persistently altered following vitreous surgery and resulted in an increased permeability index. This was due to a faster and increased accessibility of fluorescein to the vitreous, while the elimination from the vitreous was unaffected. These findings were in contrast to the anterior chamber, where an immediate affection of the blood-aqueous barrier was followed by complete restoration 1 month after surgery. The anterior chamber was examined with the same technique as the vitreous. The anterior chamber permeability index was significantly increased compared to the vitreous in normal subjects. This point towards a less efficient barrier surrounding this part of the eye. With increasing degree of retinopathy, the permeability index obtained from the anterior chamber increased. This index was found increased in all groups of diabetic subjects, compared to the permeability index obtained from the vitreous. The penetration rate constant of fluorescein (Kin) to the anterior chamber was of the same order of magnitude as the initial disposition rate constant (α) for plasma fluorescein, while the rate of elimination from the anterior chamber (Kout) was of same order of magnitude as the terminal decay in plasma. These data point towards a close relation between the anterior chamber and the central compartment. The penetration rate constant for fluorescein (Kin) to the anterior chamber was of the same order of magnitude as the initial disposition rate constant in plasma in all groups of diabetic subjects. However, the elimination rate constant obtained from the anterior chamber decreased significantly with increasing degree of retinopathy and was significantly below the terminal plasma fluorescein disposition rate (β) in both background and proliferative diabetic retinopathy. This point towards an increasing ocular restriction of drug elimination from the anterior part of the eye with increasing degree of retinopathy. Drug exchange across the ocular barriers in the anterior and posterior part of the eye seems to vary. In the posterior part of the eye, the penetration rate constant seems to influence the permeability index, because these parameters are statistically correlated. However, in the anterior part of the eye, the elimination influence the permeability index in an inversely correlated way.