V.A. Alder

Diabetic retinopathy: early functional changes.

1. The present review reports some of the earliest physiological changes that occur in the diabetic retina prior to any clinical or anatomical changes in an animal model of diabetes.

Vitreal and retinal oxygenation

This paper reports the results of experiments carried out to understand the oxygenation of the normal retina in response to alterations in physiological conditions such as inspired oxygen concentration, elevated IOP, luminance changes and occlusion of the retinal circulation. Measurements of vitreal and intraretinal PO2 in vivo in the cat using oxygen-sensitive microelectrodes demonstrated that large PO2 gradients were set up preretinally and that arterial, venous and tissue PO2 increased when inspired PO2 was raised to 100% O2. Results showed that with an occluded retinal circulation, it was possible to oxygenate fully the whole retina with oxygen supplied from a hyperoxic choroidal circulation. With alterations in background luminance from photopic to scotopic, preretinal PO2 was unaffected for air breathing, whereas for 100% O2, breathing vitreal PO2 fell quickly and reversibly on a switch from photopic to scotopic conditions, reflecting an increase in retinal oxygen consumption in a dark environment. During acute rises of IOP, the PO2 at the choriocapillaris fell and an anoxic region developed in the middle retinal layers. The inner retina was relatively resistant to rises in IOP. The implications of these data for autoregulation of the retinal circulation are discussed.

Choroidal blood flow measured in the dog eye in vivo and in vitro by local hydrogen clearance polarography : Validation of a technique and response to raised intraocular pressure

Local choroidal flow has been measured using hydrogen (H2) clearance polarography in dog eyes in vivo and in vitro. The accuracy and reproducibility of the method was checked in the eye in vitro by demonstrating a linear relationship between flow measurements obtained by an absolute method and those obtained by the H2 clearance technique. The H2 electrode was placed through a scleral window and its current response to a bolus injection of H2-saturated saline into a proximal artery was measured. The clearance of H2 from the choroidal circulation in vitro was monoexponential and the reproducibility of the measurements at the same site and in the same eye was good. In vivo the technique was used to measure alterations in local choroidal blood flow produced by acute changes of intraocular pressure. A linear relationship between perfusion pressure and choroidal blood flow was obtained. H2 clearance polarography is shown to be an accurate and reproducible method for determining choroidal flow, with the advantage that localized repeated measurements of flow can be made.

Continued progression of retinopathy despite spontaneous recovery to normoglycemia in a long-term study of streptozotocin-induced diabetes in rats

Abstract · Background: This long-term (2.3 years) study determines the temporal relationship between systemic glucose levels and the progression of diabetic retinopathy during the natural course of streptozotocin-induced diabetes in rats. · Methods: Of 367 rats, 200 were randomly assigned into a group injected with streptozotocin (50 mg.kg–1) and 167 into a control group. Subsets of the rats were killed at 6, 28, 40, 65, 90 and >100 weeks post induction to allow the severity of retinopathy to be assessed quantitatively and qualitatively by trypsin digests of the retinal vasculature. Concurrently blood glucose, body weight and death rate were monitored. · Results: Three glycemic phases were observed in the streptozotocin rats. In phase 1 (0 to 36–40 weeks) hyperglycemia was established and maintained. In phase 2 (36–40 to 84–90 weeks) normoglycemia was restored, and maintained during phase 3 (84–90 to 120 weeks). Control rats were normoglycemic throughout. The retinal microangiopathy was marked at 28 weeks during phase 1, developed more slowly in phase 2 and continued to worsen with loss of capillaries in all retinas and saccular microaneurysms present in 50% of retinas in phase 3. Cumulative death rate in streptozotocin rats also followed three phases, with maximum vulnerability occurring between 28 and 40 weeks. Body weight was significantly lower in streptozotocin rats throughout, increasing slowly in phase 1, then more rapidly during and after spontaneous glycemic recovery. · Conclusion: The worsening retinopathy, despite sustained recovery to normoglycemia, implies that good glucose control alone does not stop the progression of the retinal microangiopathy at this late stage.

Altered vasoactivity in the early diabetic eye : measured in the isolated perfused rat eye

The effect of 4 weeks streptozotocin-induced diabetes on ocular vascular resistance responses to noradrenalin (NA), adrenalin (A), phenylephrine (PHE), isoproterenol (ISOP), prostaglandin F2 alpha (PGF2 alpha). 5-hydroxytryptamine (5-HT) and angiotensin II (ANG II), was determined using a newly-developed, isolated, arterially-perfused rat eye preparation, by comparing responses from control and diabetic eyes. After extensive preliminary experiments to establish optimum parameters, the ophthalmic artery of enucleated control and diabetic rat eyes was cannulated and the retinal and uveal vasculature perfused at a constant flow with Na(+)-Krebs solution after streptozotocin-induced diabetes had been established for 4 weeks. The eyes were maintained in an environment-controlled organ bath. Perfusion pressure was monitored as increasing log M concentrations of agonists were added to the perfusate. Total ocular resistance could be calculated from knowledge of flow and pressure. In control eyes, NA, A, PHE, PGF2 alpha, and 5-HT all produced dose-dependent increases in total vascular resistance, with the following order of potency: NA = A > 5-HT > PHE = PGF2 alpha at 10(-4) M. The ocular circulation was not sensitive to isoproterenol and angiotensin II. In diabetic eyes responses to NA, A, PGF2 alpha and 5-HT were altered. Diabetic responses to NA and A had lower thresholds with larger resistance increases at low concentrations. However, the rate of increase in resistance with concentration was more gradual in diabetic eyes so that at 10(-4) M control responses were larger. Diabetic resistance responses to PGF2 alpha had the same threshold as in control eyes, but were greater in magnitude with an earlier peak at 10(-4) M. In contrast diabetic resistance responses to 5-HT were reduced, peaked at a lower resistance at 10(-4) M, but had the same threshold as those in the control eye. Basal vascular resistances in control: 3.14 +/- 0.32 mmHg min microliter-1 (n = 28), and diabetic eyes: 3.44 +/- 0.19, mmHg min microliter-1 (n = 36), were not significantly different. Vasoactivity in the early diabetic eye is disturbed with the effective balance between different agonists altered in favour of catecholamines at physiological concentrations. This may be related to the early changes in blood flow and oxygen distribution already reported in the rat eye, as well as changes to autonomic function. The isolated perfused rat eye is a valuable technique for investigating such vascular reactivity in animal models of retinal disease.

Systemic and Ocular Vascular Roles of the Antiglaucoma Agents b-Adrenergic Antagonists and Ca2+ Entry Blockers ☆

This review addresses whether the antiglaucoma agents beta-adrenergic antagonists and Ca2+ entry blockers cause vasoactive effects in the retinal and other ocular vasculatures, as they do in other tissues. The potent vasodilating effects of Ca2+ entry blockers on ocular vessels have recently been demonstrated in in vivo and in vitro studies, implying that the maintenance of ocular vascular tone relies almost exclusively on extracellular Ca2+. Ca2+ entry blockers may potentially play a role in relaxing the retinal, long posterior ciliary, and ophthalmociliary arteries to improve the ocular circulation in vascular diseases in which there is considerable vascular tone present. The beta-adrenergic antagonists are discussed with reference to their antihypertensive role, their effect on other vascular beds, and finally what is known of their effect in the ocular vasculature. The emerging evidence that particular selective beta-adrenergic antagonists, such as betaxolol, are also potent Ca2+ channel entry blockers in other vascular beds is presented. Betaxolol has been shown to induce vasodilatation in the retinal and other ocular vascular beds, although studies have shown that beta1-adrenergic receptors are sparse in these vascular beds. This implies that an alternative mechanism must be responsible for betaxolol-induced vasodilatation. Evidence is presented that betaxolol vasodilates via its potent Ca2+ channel entry blocking properties, and its potency and ability to vasodilate are compared with those of nimodipine and timolol, as well as with those of other Ca2+ channel entry blockers. Important areas for future research in this area are discussed.

Pharmacological and mechanical heterogeneity of cat isolated ophthalmociliary artery

The mechanical, histological and pharmacological properties of the isolated cat ophthalmociliary artery are reported and compared for passively stretched ring segments from three locations on the artery. Ring segments were mounted in a modified myograph system after the method of Hogestatt, Andersson and Edvinsson (1983, Acta Physiol Scand. 117, 49-61) and comparative length tension curves were measured. The most distal segment (DOA), nearest the eye, developed by far the smallest maximal tension compared to the middle segment (MOA) and the most proximal segment (POA), despite the fact that lumen diameters were similar. All segments displayed a phasic and tonic component in response to passive stretching and during activation with K(+)-Krebs. Cumulative contractile dose-response curves were measured for nine agonists: histamine (HIS), 5-hydroxytryptamine (5-HT), dopamine (DOPA), adrenaline (A), noradrenaline (NA), tyramine (TYR), phenylephrine (PHE), isoproterenol (ISOP) and xylazine (XYL) and the maximum active tension developed at a concentration of 10(-3) M was compared with that for 0.124 M K(+)-Krebs. The order of contraction magnitude in the three segments was: POA, HIS much greater than Kmax much greater than NA greater than A greater than DOPA greater than PH = TYR greater than 5-HT = ISOP greater than XYL; MOA, Kmax much greater than NA much greater than A = HIS much greater than DOPA greater than ISOP = PHE greater than TYR = 5-HT greater than XYL; DOA, Kmax much greater than NA greater than A greater than HIS = DOPA greater than ISOP greater than PHE = 5-HT greater than XYL = TYR. Pre-activation did not produce relaxation in response to histamine and isoproterenol. Testing with specific alpha 1 and beta antagonists supported the presence of alpha 1 receptors but not beta receptors. It is concluded that functional HIS, alpha 1 and 5-HT receptors are present and that the ophthalmociliary artery is heterogeneous in its response to HIS, with the POA segment producing the largest responses. The magnitude of the contractile responses to A and NA grew with increasing passive tension, whereas the response to histamine was independent of passive tension.

Changes in vitreal oxygen tension distribution in the streptozotocin diabetic rat

SummaryMeasurements of vitreal oxygen tension have been made for the first time in the streptozotocin-induced diabetic rat eye. A total of 36 Sprague Dawley rats were divided into a control (n=18) and streptozotocin injected group (n=18), and after 5–6 weeks of established hyperglycaemia, an acute experiment was performed in which vitreal oxygen tension profiles were determined with oxygen sensitive microelectrodes. The control rats had significant oxygen tension gradients in the vitreous close to retinal arteries with relatively flat oxygen tension profiles close to retinal veins and intermediate regions. All control rats had a substantial arteriovenous oxygen tension difference when measurements were made on retinal arteries and veins. In contrast the oxygen tension profiles measured in the vitreous of streptozotocin rats showed markedly reduced oxygen gradients in the vicinity of retinal arteries and a smaller arteriovenous oxygen tension difference. In both groups of rats, for distances of 500 μm and greater from the retina (mid vitreous) a plateau oxygen tension value was observed. No significant difference was found in this mean mid vitreous value between the control rats and diabetic rats under the same systemic conditions. We conclude that there are significant changes in oxygen tension near retinal arteries in streptozotocin-induced diabetes before any histopathological changes are evident.

Intraretinal oxygen tension in the rat eye

AbstractOxygen-sensitive microelectrodes were used to measure the intraretinal oxygen distribution in the rat during both air and oxygen breathing. During air breathing the average PO2 profile obtained showed a value of 19 ± 2.5 mmHg (n = 36) at the internal limiting membrane (ILM). As the penetration depth increased, a minimal PO2 of 8.2±2.4 mmHg was attained at 50% of the retinal depth, from which point the PO2 increased to a peak of 33 ± 3.3 mmHg at 100% of the retinal depth at Bruchs membrane. During oxygen ventilation, all intraretinal oxygen tensions were increased and the profile became almost monotonic, increasing from 36.1 ±3.3 mmHg (n = 9) at the ILM to 134.3 ± 6.7 mmHg at 100% of the retinal depth. These results are qualitatively similar to those found in the cat, and they demonstrate the importance of oxygen delivery by both the retinal and the choroidal circulation during normoxia, as well as the inability of vascular autoregulation to maintain a constant retinal P02 during hyperoxic ventilation. Offprint requests to: S. Cringle

Vitreal oxygen tension measurements in the rat eye.

Oxygen tension measurements are presented for the first time for the vitreous body of the rat eye. Oxygen-sensitive microelectrodes were used to measure vitreal PO2 profiles for two breathing conditions: 20 and 100% O2. Results show that for ventilation with 20% O2 most of the vitreous body has a constant PO2 (mid vitreous PO2 of 22.6 +/- 1.9 mmHg (+/- S.E.; n = 6), except for a layer about 200 microns thick at the retinal surface where significant PO2 gradients exist, especially in the region of retinal arteries. Midvitreous PO2 is similar to the PO2 at retinal veins. All vitreal PO2 values increase if ventilatory oxygen is raised to 100% O2. After equilibration, the main body of the vitreous again has an almost constant PO2, a factor of 3.75 +/- 0.40 (n = 10), higher than for 20% O2 ventilation except for a region about 400 microns thick adjacent to the internal limiting membrane where the local geometry of the retinal circulation dictates PO2 distribution. When the electrode was oriented to make contact with the retina at different distances from the optic disk, venous PO2 was observed to decrease in more peripheral locations. This work opens the possibility of investigating the role of hypoxia in rat models of retinal vascular disease.