Clinical assessment of the macula by retinal topography and thickness mapping.

PURPOSE: To report a quantitative and objective method for assessing pathologic alterations in retinal structures to improve the evaluation of macular diseases. METHODS: We used a system based on the scanning retinal thickness analyzer to generate serial optical section images of the retina and provide mapping of the retinal topography and thickness in a normal subject and in patients with representative maculopathies including traumatic macular hole, central serous chorioretinopathy, branch retinal vein occlusion, diabetic macular edema, and retinal pigment epithelial detachment. RESULTS: The effectiveness of the system in imaging both the vitreoretinal and chorioretinal interfaces was confirmed in the normal subject and in patients with various maculopathies. Mapping of retinal topography and thickness in a normal eye correlated well with normal anatomy, delineating the foveal depression clearly. The retinal thickness map in a patient with diabetic macular edema showed thickening of the retina and absence of a foveal depression. The patients with central serous chorioretinopathy and branch retinal vein occlusion had an elevated vitreoretinal surface. Conversely, the patient with retinal pigment epithelial detachments had a relatively flat vitreoretinal interface but an irregularly elevated chorioretinal surface. CONCLUSION: Quantitative mapping of retinal topography and thickness is a promising tool that may improve evaluation of macular diseases.

Effect of U74006F (Tirilazad Mesylate) in rabbit eyes after argon laser trabeculoplasty.

A new free radical scavenger, U74006F (Upjohn Company, Kalamazoo, MI, USA) was studied to determine if it would have a beneficial effect on preventing intraocular pressure (IOP) spikes and reducing blood-aqueous barrier (BAB) breakdown in rabbits after argon laser trabeculoplasty (ALT). The post-ALT IOP averaged 24.7 mm Hg (+/- 3.8 mm Hg; N = 6) and 30.3 mm Hg (+/- 6.9 mm Hg; N = 7) at 1 hour (p = 0.097) and 22.7 mm Hg (+/- 1.9) and 27.3 mm Hg (+/- 7.0 mm Hg) at 3 hours in the U74006F and vehicle groups (p = 0.137), respectively. By 48 hours, the IOP averaged 12.8 +/- 2.2 mm Hg in the U74006F group and 12.4 +/- 2.4 mm Hg in the vehicle group (P = 0.757). The aqueous fluorescein permeability (AFP) between lasered and unlasered eyes was determined for the U74006F and vehicle-treated groups by aqueous fluorophotometry. Prelaser or baseline AFP was determined for the U74006F (-0.236 +/- 0.236) and vehicle (-0.145 +/- 0.237) groups (p = 0.505). At day 2 posttreatment, the U74006F group and the placebo group both reported similar AFP of 5.109 +/- 4.831 and 5.680 +/- 4.280 (p = 0.827), respectively. At one week post-ALT, AFP for the U74006F group (0.109 +/- 0.367) had returned close to baseline, while that of the vehicle group (0.426 +/- 0.511) was still quite elevated (p = 0.220). U74006F appears to be beneficial in the prevention of acute elevation of IOP following ALT but not in the prevention of Blood-Aqueous Barrier (BAB) breakdown.(ABSTRACT TRUNCATED AT 250 WORDS)

Reduction of macular oedema by acetazolamide in patients with chronic iridocyclitis: a randomised prospective crossover study.

Thirty patients with cystoid macular oedema secondary to chronic iridocyclitis were enrolled in a two period, prospective, randomised, double masked, crossover study that compared sustained release acetazolamide (500 mg twice a day) with a placebo to measure the effects on the reduction of cystoid macular oedema and improvement of visual acuity. All patients were treated for 1 month with either acetazolamide or placebo, received no treatment for 1 month, and were then treated for 1 month with the other medication. Statistically significant improvement in visual acuity was seen at 14 and 28 days in the treated patients. No improvement was seen when patients received placebo. Improved visual acuity was not associated with race or sex. However, younger patients (under age 55 years) were more likely to benefit from treatment. Results of vitreous fluorophotometry, obtained at baseline and 4 weeks, demonstrated an improvement in posterior vitreous penetration ratios and mid vitreous penetration ratios after treatment with acetazolamide but not with placebo.

Retinal thickness analysis for quantitative assessment of diabetic macular edema.

Diabetic macular edema is a major cause of vision loss and is evaluated with qualitative or semiquantitative techniques. A new quantitative method for assessment of macular edema using retinal thickness analysis was applied to 19 patients with diabetic macular edema. Foveal thickening was frequently coupled with poor visual acuity. Slit-lamp biomicroscopy and stereophotography detected 80% and 78% of local areas of thickening, respectively, but failed to detect locations with average thicknesses of 1.5 and 1.6 times normal, respectively. Fluorescein leakage on angiography was generally associated with retinal thickening, but locations with similar degrees of leakage had widely varying retinal thickening. Fluorescein leakage in the posterior vitreous correlated poorly with the degree of foveal thickening. These results indicate that quantitative measurement of retinal thickness may become useful in the management of diabetic patients with macular edema.

Limitation of retinal injury by vitreoperfusion initiated after onset of ischemia.

We tested whether vitreoperfusion, a new method of perfusing the vitreous cavity with solutions containing nutrients, can limit retinal injury if initiated after the onset of ischemia. Severe bilateral ocular ischemia was induced in cats with healed lensectomy-vitrectomy wounds; 30, 60, 90, or 120 minutes later, one eye from each of 18 cats underwent vitreoperfusion while the ischemia continued for 120 minutes. The other eye simultaneously underwent either continued untreated ischemia or reinstated circulation. The histopathologic abnormalities evident after 8 days depended on the duration of ischemia. Reinstated circulation yielded less retinal damage than continued ischemia. Nine additional cats underwent bilateral ischemia for at least 210 minutes. Vitreoperfusion was initiated in one eye after 30 minutes. In each cat, the vitreoperfused eye fared significantly better as observed histopathologically and electroretinographically. We believe that no other treatment has similarly limited retinal injury in vivo when initiated so long after total ocular ischemia has developed.

Vitreous fluorophotometry in patients with Best's macular dystrophy.

Ten patients with Best's macular dystrophy were examined with vitreous fluorophotometry and results were compared with a normal population. Seventeen of the 20 affected eyes demonstrated an intact blood-retinal barrier with normal inward permeability of fluorescein dye. Despite diffuse functional impairment of the retinal pigment epithelium (RPE), determined by electro-oculography, as well as accumulation of a lipofuscin or lipofuscin-like substance within virtually all RPE cells, the blood-retinal barrier function of these cells remained intact as determined by clinical fluorophotometry. Similar findings have previously been noted in another hereditary retinal disorder (fundus flavimaculatus) in which a lipofuscin-like substance also accumulates diffusely within RPE cells.

Vitreous fluorophotometry in carriers of choroideremia and X-linked retinitis pigmentosa.

The status of the blood-retinal barrier (BRB) in carriers of choroideremia and X-linked retinitis pigmentosa (XLRP) was determined by vitreous fluorophotometry (VF) and compared with that in female control subjects. Electroretinographic (ERG) amplitudes were measured to determine the overall functional integrity of retinal rods and cones. Comparison of the VF results showed an abnormal BRB in at least some carriers of XLRP, particularly those with peripheral fundus pigmentary changes, but not in carriers of choroideremia with even moderately extensive pigmentary changes. The abnormal BRB in XLRP carriers, with or without peripheral fundus pigmentary changes, was associated with at least moderate to moderately extensive reduction in scotopic ERG amplitudes, while the normal VF results in choroideremia carriers were associated with normal scotopic ERG amplitudes. However, in XLRP carriers, mild to modest reductions in ERG scotopic responses were seen in the presence of normal VF findings.

Posterior sub-Tenon's injections of corticosteroids in uveitis patients with cystoid macular edema.

Cystoid macular edema (CME) is a major cause of visual impairment and is thought to be due to abnormal perifoveal capillary permeability. Posterior sub-Tenon's corticosteroid injections are used to improve the visual acuity in CME, although their mechanism of action is uncertain. In this study, visual acuity, blood retinal barrier (BRB) permeability, and fluorescein angiograms were recorded immediately before and one and four weeks after the administration of steroid injections. Ten patients (12 treated eyes) with CME secondary to uveitis were studied. Visual improvement, defined as an increase in at least two lines of Snellen visual acuity, was seen in half of the treated eyes. In some patients, these improvements were not directly related to changes in the BRB permeability or the amount of macular fluid. Posterior sub-Tenon's corticosteroid injections do not consistently affect blood retinal barrier permeability.

Vitreous fluorophotometry in patients with cataract surgery.

Cystoid macular edema frequently complicates recovery from cataract surgery. Our understanding of this disease would be enhanced by more sensitive and quantitative methods to evaluate it. We performed vitreous fluorophotometry in 66 nondiabetic patients who had recently undergone cataract surgery. Eyes with definite macular edema on fluorescein angiography were ranked according to the severity of the leakage. The rank of the leakage correlated significantly with the vitreous fluorophotometry values (P less than 0.01). The mean fluorophotometry result in the eyes with cystoid macular edema was greater than that in the phakic, control group (P less than 0.01). However, eyes with minimal or no leakage visible on fluorescein angiography had fluorophotometry values indistinguishable from normal. The quantitative vitreous fluorophotometry results provided an indicator of cystoid macular edema that may prove useful in following this problem and evaluating therapy for it.

Ocular fluorophotometry in the normal- and diabetic monkey.

Normal- and diabetic rhesus monkeys without retinopathy demonstrable by ophthalmoscopy or fluorescein angiography were examined with ocular fluorophotometry to detect alterations in their blood-ocular barriers. All vitreous fluorophotometry values were corrected for fluorescence attributable to background levels and then normalized to a blood fluorescein level of 10 micrograms ml-1. Reproducibility studies demonstrated an average coefficient of variation of 0.17 for all animals combined. Insulin-dependent monkeys, both pancreatectomized and streptozotocin-treated, demonstrated significantly higher posterior vitreous fluorescence levels than either control animals or monkeys treated with streptozotocin that were not insulin-dependent. These results cannot be attributed to differences in fluorescein binding or to vitreous abnormalities. However, 14 out of 24 (58%) of the insulin-dependent animals exhibited posterior vitreous fluorescence values within two standard deviations of the control mean. No correlation was apparent between the vitreous values and age or duration of treatment. No difference in anterior chamber concentrations was found between groups after correction. Our results indicate that alterations in blood-retinal barrier can occur in insulin-dependent diabetic monkeys before development of retinopathy.

Blood-retinal barrier permeability to carboxyfluorescein and fluorescein in monkeys.

Lipid solubility is a major determinant of permeability across the blood-brain barrier, to which the blood-retinal barrier (BRB) has many similarities. Carboxyfluorescein is a dye with about 1/1000 the lipid solubility of fluorescein, but their molecular sizes and spectral characteristics are similar. We studied the importance of lipid solubility in BRB permeability by comparing the BRB permeabilities to these two dyes. Dye in the vitreous and plasma of four monkeys was measured by fluorophotometry. The estimated inward permeability coefficients (Pin) were 11 +/- 7.4 X 10(-6) cm/min (mean and SD) for carboxyfluorescein and 21 +/- 5.9 X 10(-6) cm/min for fluorescein. The ratio of the means was 1/1.9, far from the expected 1/1000. This finding suggests that the BRB does not function as a continuous lipid membrane and that other factors are more important determinants of permeability for these dyes than lipid solubility.

The effect of pH on the transfer of fluorescein across the blood-retinal barrier.

The blood-retinal barrier (BRB) might be governed by the same permeability principles as the blood-brain barrier (BBB). For a weak acid like fluorescein, BRB permeability would be proportional to its pH-dependent lipid solubility, according to the pH partition hypothesis. A range of metabolic acidosis was produced in 20 rats by the oral administration of NH4Cl; six additional rats received normal saline. Four hours later, vitreous fluorophotometry, venous fluorescein values, and arterial pH were measured. Significant linear relationships were found between vitreous fluorophotometry readings and blood hydrogen ion concentrations (p less than 0.025) and plasma fluorescein concentrations (p less than 0.05). According to the linear relationship, changing the pH from 7.4 to 7.3 or 6.9 would result in an increase in vitreous fluorophotometry reading of 8.5 or 72%, respectively. Since the pH partition hypothesis predicts values of 52 or 640% our results suggest that the BRB conforms less to the hypothesis than does the BBB. Furthermore, although pH changes of a magnitude able to influence vitreous fluorophotometry readings substantially may occur under experimental conditions in animals, they are unlikely to occur in ambulatory human patients.

The performance of a new commercial ocular fluorophotometer in the clinical environment.

Studies using the Fluorotron Master fluorophotometer in vitro and in vivo are presented. The in vitro tests include lower limit of detection and error of measurement, and the in vivo tests include spread function, axial resolution, and reproducibility. Results show that many technical problems have been resolved by this system. Furthermore, this system can be satisfactory handled by technical personnel after short periods of training.

Pathophysiology of the blood-retinal barrier in experimental diabetes. Vitreous fluorophotometry using carboxyfluorescein and fluorescein.

Carboxyfluorescein resembles fluorescein in size and spectral characteristics but is much less lipid soluble. Both dyes were used to differentiate between two groups of factors that influence penetration across the blood-retinal barrier: (1) factors that depend on lipid solubility, such as the area of the barrier, and (2) factors independent of lipid solubility, such as opened intercellular junctions or necrotic cells. Vitreous fluorophotometry was performed on normal and diabetic rats after injection of either dye. After the results were adjusted for sources of error, midvitreous-plasma dye ratios for carboxyfluorescein and fluorescein were of the same order of magnitude in normal rats. Ratios for both dyes increased in diabetic rats, and the increases were similar in magnitude. Our results suggest that lipid solubility contributes little to inward transport of these dyes in both the normal and diabetic states.

Outward transport of fluorescein from the vitreous in normal human subjects.

The permeability of the blood-vitreous barrier to fluorescein passing out of the vitreous does not necessarily equal the permeability to fluorescein passing into it. We calculated the ratio between the outward and inward permeability coefficients of the blood-vitreous barrier in eight normal men who ingested 3-g of sodium fluorescein. The calculation was based on the ratio between the serum free fluorescein and the vitreous fluorescein concentrations (as determined by fluorophotometry) when the net transport across the barrier was zero. The outward permeability to fluorescein was 31 +/- 18 times (mean +/- SD) the inward permeability. To our knowledge, this article provides the first direct evidence for a specialized transport mechanism in humans whereby fluorescein is removed from the vitreous into the blood. The malfunction of this process may be important in human diseases. Pharmacologic manipulation of this process may be possible.

Vitreous fluorophotometry in the alloxan- and streptozocin-treated rat.

Since serum fluorescein levels are altered in various experimental states, we examined the relationship between serum and vitreous fluorescein levels. We also injected streptozocin, alloxan, and alloxan with glucose pretreatment do determine whether alterations in the blood-retinal barrier are directly attributable to drug toxicity. When corrected for serum levels, one- and two-hour vitreous fluorescein levels increased above prediabetic values two days after streptozocin or alloxan administration; two-hour readings were higher. Rats treated with alloxan plus glucose did not become diabetic or show elevated vitreous fluorescein levels. Insulin treatment of alloxan-induced diabetic rats resulted in normal vitreous readings without normoglycemia. These results suggest that vitreous readings should be corrected for serum levels and that observations at two hours could be more sensitive than at one hour. Furthermore, the observed alterations in blood-retinal barrier function are not attributed to drug toxicity.