Comparative toxicity of intraocular irrigating solutions on the corneal endothelium.

Isolated rabbit and human corneas were perfused in vitro with Plasma-lyte 148 solution and with a glutathione bicarbonate Ringer's solution. The corneal swelling rate and ultrastructure were compared to corneas perfused with three commonly used intraocular irrigating solutions. Corneas perfused with Plasma-lyte 148 swelled at a rate of 47 mu/gr and the endothelial cells separated from each other and showed extensive damage after three hours of perfusion. By comparison, corneas perfused with 0.9% NaCl increased in thickness by 98 mu/hr, lactated Ringer's by 39 mu/hr, balanced salt solution by 24 mu/hr, and glutathione bicarbonate Ringer's solution by 4 mu/hr. These results indicate that endothelial cell damage and increased corneal thickness observed during perfusion was related to the incomplete composition of 0.9% NaCl, Plasma-lyte 148, and lactated Ringer's and that endothelial cell damage can be prevented if the intraocular irrigating solution contains concentrations of inorganic and organic constituents that are similar to those in aqueous humor.     

Contact lens-induced edema in vitro. Pharmacology and metabolic considerations

The effects of physiologic and pharmacologic manipulations on contact lens-induced edema were studied. In isolated superfused rabbit corneas bathed in Ringer's solution and covered with large-diameter polymethylmethacrylate (PMMA) lenses, corneal swelling rates of 17-26 microns/hr (versus -5-5 microns/hr in paired controls) were observed. Neither the calcium antagonist diltiazem (10(-4) M), the glucocorticoid dexamethasone (10(-7) M), the glucose substitute fructose (20 mM), nor 0.5 mM adenosine and 0.3 mM reduced glutathione mitigated the edema. Lens-induced edema was 25 microns/hr in corneas bathed at pH 8.2 and decreased to 9 microns/hr at pH 7.0. In corneas without lenses, however, decreasing the pH from 7.4-7.0 caused significant swelling (P less than 0.05). The pyruvate dehydrogenase stimulant sodium dichloroacetate (3.2 mM) on the tears side ameliorated the edema, and its congener, 3.2 mM 2-chloropropionate, was less effective. These latter agents are known to relieve lactic acidosis systemically and had no significant effect on corneas without lenses. In tissues bathed with 20 mM lactate Ringer's, normal thickness was maintained in both control and PMMA-treated corneas throughout the 3-hr period. These findings suggest that the contact lens-induced edema does not involve the acute cytotoxic mechanisms seen in severe tissue ischemia or hypoxia. The edema appears to result in part from acidosis but mainly from stromal lactate accumulation.     

Effects of sodium lactate on isolated rabbit corneas

Corneal stromal lactate accumulation may result from epithelial hypoxia and contact lens wear, but the possible corneal toxicity of lactate has not been reported. Isolated superfused whole rabbit corneas were examined for thickness changes during exposure to neutral sodium lactate (NaL) or excess sodium chloride (NaCl) in Krebs-bicarbonate Ringer's solution for a 3-hr period. Placed in the tears side bath, 5 mM NaL significantly thinned corneas (swelling rates of 1 +/- 1 micron/hr in Ringer's controls vs -11 +/- 1 micron/hr in lactate-treated corneas; mean +/- SD). Excesses of 5 mM NaCl had essentially identical effects (0 +/- 1 micron/hr in controls vs -13 +/- 3 micron/hr in experimentals). When placed on the aqueous side of normal-thickness corneas, neither 20 mM NaL nor 20 mM excess NaCl affected corneal thickness, but both solutions stimulated endothelium-mediated deswelling in preswollen deepithelialized corneas. When "loaded" into the stroma of deepithelialized corneas, Ringer containing 20 mM lactate caused more swelling than Ringer's alone (491 +/- 18 microns in controls vs 558 +/- 20 microns in loaded corneas; mean +/- SEM). A similar swelling occurred when 20 mM excess NaCl was loaded into the stroma (483 +/- 15 vs 565 +/- 20 microns in controls and loaded corneas, respectively), due to fluid uptake into the hypertonic stroma across the endothelium from the aqueous side (Ringer's) bath. Corneas both loaded and superfused with either NaL or excess NaCl swelled and subsequently deswelled similar to controls swollen and superfused in Ringer's.(ABSTRACT TRUNCATED AT 250 WORDS)     

A system for long-term corneal perfusion

Seventy-two human corneas were maintained in a perfusion system at 37 degrees C and 18 mm Hg intracameral pressure for 1 to 3 weeks. Corneal thickness, which was initially greater than normal because the enucleated eyes were kept at 4 degrees C before excision of the corneas, decreased slowly during the period of incubation. Endothelial removal or perfusion with ouabain (10(-4) M) induced irreversible stromal swelling. Cooling to 4 degrees C for 8 hr during perfusion caused stromal swelling that disappeared after rewarming to 37 degrees C; less stromal swelling occurred with cooling after 3 weeks of perfusion than after 3 days. No enlargement of central endothelial cells was noted in most corneas by serial specular microscopy. Electron microscopy demonstrated reversal of postmortem changes and maintenance of normal intracellular ultrastructure for 3 weeks. This system for long-term corneal perfusion will allow controlled studies of the effects of new methods of corneal preservation and other perturbations upon the corneal endothelium in situ.     

Corneal endothelial junctions and the effect of ouabain

Paired rabbit corneas were perfused in vitro for endothelial permeability (Pac) determination with glutathione bicarbonate Ringer's solution (GBR) and GBR plus ouabain (10(-4) M). Results indicated no difference in Pac between the two groups (3.39 vs 3.67, respectively) despite significantly greater stromal swelling in the group perfused with ouabain. Freeze-fracture microscopy of similarly perfused corneas revealed intact tight junctional complexes in both groups, although the tight junctional complex of perfused corneas appeared less organized than that of freshly enucleated, nonperfused controls. Gap junctions were abundant as observed in freeze-fracture replicas of GBR-perfused endothelium, and appeared to be decreased or absent in ouabain-perfused endothelium. These results indicate that corneal endothelial tight junctions are unaffected by perfusion with ouabain, whereas gap junctions appear to be lost. The permeability and freeze-fracture data reaffirms the importance of tight junctions as permeability barriers and indicates that gap junctions are not of primary importance for maintenance or control of the corneal endothelial barrier.     

Effects of irrigation solutions on corneal endothelial function.

Rabbit corneas were perfused in vitro with an irrigation solution for 90 minutes. This was followed by 6 hours of perfusion with tissue culture medium TC199 during which endothelial function was assessed by monitoring rates of swelling during a period of perfusion in the absence of bicarbonate ions, and subsequent rates of thinning when bicarbonate ions were restored to the perfusate. Corneal thickness (measured with an ultrasonic pachymeter) immediately following excision was 401 microns (SD 19, n = 23). During the 90 minute perfusion at 35 degrees C, corneas exposed to balanced salt solution (BSS), Hartmann's solution or 0.9% NaCl (all initially at room temperature) swelled, respectively, at 14 (SD 2.3, n = 4), 11 (SD 2.6, n = 4), and 70 (SD 4.3, n = 4) microns/h. Cold Hartmann's solution (initially at 4 degrees C) caused corneas to swell at 9 (SD 2.3, n = 4) microns/h. On the other hand, corneas perfused with BSS Plus thinned at 9 (SD 3.4, n = 4) microns/h and TC199 with Earle's salts had little effect on thickness. Rates of swelling and thinning during the following assessment perfusion showed no apparent effects of prior exposure to any of the irrigation solutions on the barrier properties or pump function of the endothelium. Despite this, the increased thickness of corneas exposed initially to BSS, cold Hartmann's solution, or 0.9% NaCl was not fully reversed, even by the end of the 6 hour assessment perfusion. In contrast, the swelling observed in corneas exposed to Hartmann's solution at room temperature was reversed and these corneas had returned to their normal thickness by the end of the assessment period. All corneas, even those exposed to 0.9% NaCl, had an intact endothelial mosaic with no evidence of damage or cell loss, although morphological differences in cell shape and the appearance of cell borders were evident compared with freshly isolated cornea.     

Contact lens-induced edema in vitro. Ion transport and metabolic considerations

The relationship of contact lens-induced edema to epithelial and endothelial function was determined in isolated superfused rabbit corneas. Placement of a polymethyl methacrylate (PMMA) contact lens on the cornea caused swelling rates of 15-28 microns/hr compared to 0-6 microns/hr in paired control corneas. The edema increased with temperature (P less than 0.01). PMMA-induced swelling was significant in: 1) bicarbonate-free Ringer's solution; 2) chloride-free Ringer's; 3) 0.3 mM furosemide-treated corneas; and 4) deepithelialized corneas. The swelling did not occur in corneas with silicone oil replacing the endothelium to block fluid uptake. The effluent aqueous bathing fluid from edematous corneas did not induce edema in normoxic corneas. These studies demonstrate that contact lens-induced edema depends on metabolism, involves a significant stromal contribution, and requires fluid absorption across the endothelial layer, but is not a direct result of epithelial and endothelial ion transport inhibition.     

Physiological state of the rabbit cornea following 4 degrees C moist chamber storage

Eyes from female albino rabbits (1.9-2.3 kg) were enucleated immediately following euthanasia along with the lids and conjunctiva. The globes were moist chamber-stored (in a 4 degrees C refrigerator with the lids closed and the cornea facing downwards) for 12-500 hr. Central corneal thickness (by ultrasound) increased from 350 to approximately 650 microns within 72 hr but changed little thereafter. In the latter period, the relative fluid pressure of the globe (pneumatonography) dropped to less than 10 mmHg, aqueous bicarbonate (tCO2) levels fell to less than 5 mM and anterior chamber fluid tonicity decreased progressively (especially after 24 hr storage) to reach values of around 200 mosmol l-1 by 7 days. Storma-endothelium preparations of the corneas, after 2 hr of in vitro equilibration with a bicarbonate-Ringer solution (supplemented with glucose, adenosine and glutathione) were evaluated for their ability to undergo deturgescence under silicone oil or to pump fluid against a hydrostatic pressure of 20 cmH2O. Corneal preparations from up to 7 days storage showed rapid (60 to 135 microns hr-1) and complete deturgescence (net change in thickness of 140-180 microns) that was maintained. Thereafter, the ability to show deturgescence declined to zero by 10 days. In marked contrast, endothelial fluid pump activity (of approximately 5 microliters hr-1) was manifest for only 36 hr after which time this function rapidly declined. Most corneas stored for periods longer than 48 hr exhibited a continuous leak (of -1 to -5 microliters hr-1). The results indicate that corneal deturgescence and endothelial fluid pump function are not necessarily coupled in vitro.     

Comparison of preservative-free bupivacaine vs. lidocaine for intracameral anesthesia: a randomized clinical trial and in vitro analysis

PURPOSE: To determine whether intracameral bupivacaine hydrochloride 0.5% is as effective as lidocaine hydrochloride 1.0% in controlling discomfort of patients during phacoemulsification and posterior chamber intraocular lens implantation. In rabbits, corneal endothelial cell function, ultrastructure, and viability were evaluated after in vitro perfusion of bupivacaine 0.5%. METHODS: In a double-masked, controlled trial, 48 eyes of 48 patients with uncomplicated age-related cataract were randomly assigned to receive bupivacaine 0.5% or lidocaine 1.0% intracamerally before phacoemulsification with a posterior chamber intraocular lens. Outcome measures such as pain, visual acuity, amount of sedation, length of surgery, pupil size, intraocular pressure, corneal clarity, and anterior chamber reaction were compared. In laboratory studies, paired rabbit corneas were evaluated by endothelial cell perfusion with either bupivacaine 0.5%, bupivacaine 0.5% and glutathione bicarbonate Ringer solution in a 1:1 ratio or bupivacaine 0.5% buffered to a pH of 7.0. The paired control corneas were perfused with glutathione bicarbonate Ringer solution and rates of corneal swelling were determined. Cell ultrastructure and viability were also evaluated. RESULTS: In the randomized trial, there was no significant difference in the pain patients had during surgery or in the early or late postoperative period. No statistically significant difference was seen between the two groups in terms of pupil size, intraocular pressure, corneal edema, anterior chamber reaction, or visual acuity immediately after the operation or on postoperative day 1. Paired rabbit corneas perfused with bupivacaine 0.5% and bupivacaine 0.5% buffered to a pH of 7.0 swelled significantly (P<.001, P = .009, respectively), and had corneal endothelial cell damage. Dilution of the bupivacaine 1:1 with glutathione bicarbonate Ringer solution prevented corneal edema and damage to the corneal endothelium. Endothelial cell viability was also decreased after perfusion of bupivacaine 0.5% (P<.001). CONCLUSIONS: Clinically, bupivacaine 0.5% is as effective as lidocaine 1.0% for anesthesia during phacoemulsification and posterior chamber intraocular lens implantation. However, in vitro perfusion of bupivacaine 0.5% damaged the corneal endothelium of rabbits except when the drug was diluted 1:1 with glutathione bicarbonate Ringer solution. Surgeons who use 0.2 to 0.5 ml of intracameral bupivacaine 0.5% should be aware of its potential to cause endothelial cell damage because of its lipid solubility. The bupivacaine 0.5% should be diluted at least 1:1 with balanced salt solution before intracameral injection, followed immediately by phacoemulsification. The surgeon should ensure that the bupivacaine 0.5% is nonpreserved and packaged in single-use vials or flip-top containers.     

Effect of irrigating solution and irrigation temperature on the cornea and pupil during phacoemulsification

PURPOSE: To evaluate the effect of irrigation solution and temperature on pupil diameter, corneal endothelium, and corneal pachymetry during and after phacoemulsification. SETTING: Klinik Dardenne, Bonn, Germany. METHODS: Eighty patients who had cataract surgery by phacoemulsification were assigned to 1 of 4 cross-classified groups and had intraoperative irrigation with room-temperature or refrigerated fortified balanced salt solution (BSS Plus) or modified Ringer's solution. Pupil diameters were recorded at different stages during the surgery. Epithelial cell counts and pachymetry were determined before and 1 day after surgery. RESULTS: The solution temperature did not affect any parameter. The type of solution did not influence endothelial cell loss; however, the solution had a significant effect on corneal pachymetry 1 day postoperatively. The corneas irrigated with BSS Plus were less swollen than the corneas irrigated with Ringer's solution. By day 14, corneal thickness was equal among all groups. CONCLUSION: Long-term results were equally favorable in all 4 groups. However BSS Plus induced less short-term corneal swelling than Ringer's solution. From these findings, it appears that BSS Plus may decrease corneal risk in cases with compromised corneas or prolonged surgery.     

Loss of stromal glycosaminoglycans during corneal edema

This study tried to determine if glycosaminoglycans (GAGs) are released from the rabbit stroma during corneal edema. The GAGs of rabbit corneas were labeled in situ using anterior-chamber injections of 35S-sulfate and 3H-glucosamine. Labeled corneal pairs were excised and the endothelium perfused in vitro in the specular microscope. Edema was induced in one cornea by perfusion with a calcium-free balanced salt solution; the control cornea was perfused with glutathione bicarbonate Ringer's (GBR). Corneal thickness was measured every 15 minutes during the 3-hour perfusion period, and perfusate fractions were collected from each cornea and analyzed for the presence of GAGs. Edematous corneas swelled from 438 +/- 14.8 microns to 688 +/- 10.6 microns compared with control corneas (427 +/- 4.7 microns to 454 +/- 7.2 microns). Total 3H-glucosamine (4.00 +/- 0.68%) and 35S-sulfate (10.36 +/- 0.92%) released from the edematous corneas during perfusion exceeded that lost by control corneas (1.92 +/- 0.18% for 3H-glucosamine; 3.23 +/- 0.52% for 35S-sulfate). Enzymatic digestion studies showed the presence of keratan sulfate in the edematous perfusates. The results suggest that increased loss of radiolabeled components from edematous corneas represent a loss of stromal GAGs and possibly GAG fragments. Therefore, corneal edema involves loss of GAGs and water uptake.     

Recovery of endothelial function after vitrification of cornea at -110 degrees C

PURPOSE: To determine whether endothelial function is retained after ice-free cryopreservation of cornea by vitrification at -110 degrees C. METHODS: Rabbit corneas, mounted on support rings, were exposed to a solution containing 6.8 M propane-1,2-diol (PROH) and cooled at approximately 7 degrees C/min to -110 degrees C, which was below the glass transition temperature (T(g)) of the solution. After rewarming at approximately 12 degrees C/min and removal of the PROH, endothelial function was assessed by monitoring corneal thickness during perfusion at 34 degrees C. RESULTS: Addition and removal of 6.8 M PROH without cooling to -110 degrees C did not markedly impair endothelial function, although corneas were thicker than control samples. There was no visible crystallization of ice during cooling to -110 degrees C; but a few small, discrete sites of crystallization remote from the endothelium, were observed during warming. After removal of the PROH, corneas approximately doubled in thickness during the first 3 hours of perfusion, but they then started to thin, which suggested active control of stromal hydration by the endothelium. This was confirmed in a further set of experiments by removal of bicarbonate ions from the perfusate at this point, which resulted in further swelling at +58 +/- 2 microm/hour (SD; n = 4). Restoring bicarbonate to the perfusate halted this swelling, and the corneas then thinned at -13 +/- 2 microm/hour (n = 4). Morphologically, staining with trypan blue and alizarin red S showed an apparently intact endothelial monolayer. CONCLUSIONS: Rabbit corneal endothelium tolerated exposure to 6.8 M PROH, and endothelial function was evident after vitrification at -110 degrees C. Preliminary morphologic results with vitrified human cornea also showed retention of endothelium.     

Intraocular irrigating solutions: a comparison of Hartmann's lactated Ringer's solution,BSS and BSS Plus

Background: We evaluated the effects of Hartmann's lactated Ringer's (HLR) solution, balanced salt solution (BSS) and BSS Plus on human corneal endothelium. Methods: Paired human corneas were mounted in the in vitro specular microscope for endothelial perfusion with HLR, BSS or BSS Plus for 15, 30, 60 and 120 min. Reversal experiments with BSS Plus after initial HLR perfusion were performed. At the end of the perfusions, electron microscopy, F-actin staining of the endothelial cytoskeleton and endothelial permeability measurements were carried out. Results: Longterm perfusion (120 min) with HLR resulted in a significantly higher swelling rate than in the paired controls perfused with BSS Plus. Short-term exposure to HLR for 15, 30 and 60 min after initial BSS Plus perfusion increased the swelling rates significantly. The increased corneal swelling after HLR perfusion for 60 min was reversed by BSS Plus perfusion. Ultrastructural changes in HLR-perfused corneas included endothelial cell edema, cytoplasmic vacuolation and mitochondrial swelling. F-actin staining showed overall cytoskeletal disorganization after perfusion with HLR. Corneal endothelial permeability was higher for BSS Plus-perfused corneas than with HLR solution. Conclusion: The results suggest that the clinically observed corneal clouding during irrigation with HLR is due to endothelial cell edema and decreased endothelial pump function. However, this increased corneal swelling is reversible by perfusion with BSS Plus.     

Effect of intraocular irrigating solutions on intracellular membrane potentials and swelling rate of isolated human and rabbit cornea.

Isolated human and rabbit corneas were incubated in glutathione bicarbonate Ringer solution (GBR), balanced salt solution (BSS), or 0.9% NaCl solution. The swelling rate of human corneas was 25.5 micron/hr in GBR and significantly increased to 32.7 in BSS and 66.1 in NaCl. The epithelial intracellular potential of human cornea was constant at about 60 mV up to 5 hr of incubation in GBR and decreased continuously to 40 mV in BSS and NaCl. Endothelial cell potentials were stable for up to 3 hr of incubation in GBR or BSS and decreased from a control value of about 18 mV to 10 mV 2 hr after bathing in NaCl. Qualitatively similar data were obtained in the isolated rabbit cornea. The results demonstrate the advantage of GBR as an intraocular irrigating solution.     

Sided effects of bicarbonate on rabbit corneal endothelium in vitro

Stroma-endothelium preparations from albino rabbits were studied either in a specular microscope (and measurements made of corneal thinning under silicone oil) or mounted between two half-chambers (to permit direct measurement of endothelial fluid pump activity). Preparations were first equilibrated with bicarbonate-Ringer solutions (supplemented with glucose, adenosine and glutathione and equilibrated with 5% CO₂-air). If the preparations were equilibrated with a low (2 mM) level of bicarbonate on the stromal side and variable levels (2 to 50 mM) of bicarbonate on the endothelial side, corneal thinning increased with bicarbonate concentration while fluid pump was found to be constant from 2 to 35 mM bicarbonate. In contrast, if the preparations were equilibrated with a low (2 mM) bicarbonate level on the endothelial side and 2 to 50 mM bicarbonate on the stromal aspect, corneal thinning did not occur but fluid pump declined gradually as a function of stromal bicarbonate levels. These results contrast sharply with the significant corneal thinning and fluid pump that can be measured if both sides of the preparations are equilibrated with 35 mM bicarbonate. These results further emphasise that bicarbonate ions do not simply serve to drive an endothelium-sited fluid pump in the mammalian cornea.     

Tear potassium contributes to maintenance of corneal thickness.

Isolated rabbit corneas were bathed on their endothelial surfaces with normal Krebs bicarbonate Ringer solution, while the epithelial surfaces were bathed in a basic tear solution containing sodium and potassium. When bathed in basic tear solution alone, corneal swelling occurred at an average of 12 microns/h over a 3-hour period. Corneal swelling occurred at a rate of about 21 microns/h when the epithelial solution was switched from normal basic tear solution to an iso-osmotic K(+)-free basic tear solution. Corneal swelling then slowed, and in the final hour of a 3-hour exposure to K(+)-free tear solution, the corneas deswelled at about 10 microns/h. The data indicate that potassium is a necessary solute for the maintenance of normal corneal thickness. The results suggest that a lacrimal dysfunction that would cause a decrease in the potassium content of tears may influence corneal thickness and also suggest that the inclusion of potassium in artificial tears is important.     

Effect of HEPES buffer on corneal storage in MK medium

Rabbit corneas were stored for 7 days in either MK medium containing gentamicin or modified MK medium containing HEPES buffer, gentamicin and phenol red. Corneas stored for 7 days in modified MK medium were thicker than corneas stored in MK medium. Corneal endothelial permeability to inulin and dextran was similar following 7 days of storage in either solution. Transmission electron microscopy of corneal endothelial cells stored in either solution showed intact cell membranes and organelles. In vitro perfusion of rabbit corneas in the specular microscope with Krebs Ringer bicarbonate containing HEPES buffer swelled at 17 +/- 1 micron/h, whereas those perfused with Krebs Ringer bicarbonate alone swelled at 7 +/- micron/h. Perfusion with Krebs Ringer bicarbonate containing phenol red did not result in an increased corneal swelling rate. The work indicates that HEPES buffer has an adverse effect on corneal endothelial pumping function, and this results in corneal swelling during storage as well as during perfusion in the specular microscope. The adverse effect appears to be, at least in part, transient: however, the ultimate, long term effect of HEPES buffer on corneas stored prior to penetrating keratoplasty is not known and deserves continued investigation.     

Effect of hematoporphyrin derivative on rabbit corneal endothelial cell function and ultrastructure

Hematoporphyrin derivative (HpD) is a systemically administered photosensitizing agent that may be of value in the treatment of solid tumors. When corneal endothelial cells were perfused in the specular microscope with HpD and exposed to a 25-W incandescent light at 5 cm (5.5 mW/cm2) there was anatomic disruption of corneal endothelial cells and swelling of the corneal stroma. Perfusion with 0.2 microliter/ml (1.0 microgram/ml) HpD and 5 min exposure to light resulted in a corneal swelling of 71 +/- 4 microns after 3 hr, whereas perfusion with 0.2 microliter/ml HpD and a 1-min exposure to light resulted in a corneal swelling of 36 +/- 4 microns after 3 hr. Perfusion with 0.2 microliter/ml HpD with no light exposure resulted in a corneal swelling of 22 +/- 4 microns after 3 hr. Inclusion of 100 micrograms/ml catalase in the perfusion solution resulted in a significant 38% reduction of the corneal swelling. The inclusion of either 100 micrograms/ml superoxide dismutase, 15 mM D-mannitol, 5 mM ascorbic acid, 1/4% DMSO, 50 microns EDTA, 50 microns DETAPAC, 10 mM L-histidine, or 1 mM sodium azide did not modify the corneal swelling induced by the photosensitization reaction. Perfusion of corneal endothelial cells with 2 microliters/ml (10 micrograms/ml) HpD and exposure to 25-W incandescent light for 5 min resulted in swelling of mitochondria, the appearance of vacuoles in the cytoplasm, and rapid corneal swelling. The data suggests that corneal endothelial cells can be damaged by hydrogen peroxide generated by the dismutation of superoxide anion produced during the photoreaction. Superoxide anion itself and hydroxyl-free radical do not appear to participate in causing the endothelial cell damage. The role of singlet oxygen remains somewhat unclear. The data suggests that further in vivo studies should be performed to delineate precautions that should be taken to protect the corneal endothelium during photoradiation therapy.     

Effect of oxidized glutathione on the barrier function of the corneal endothelium

Effects of glutathione on the corneal endothelium were reexamined. Four kinds of solutions were made: oxidized glutathione (GSSG) was added to a basic solution which does not contain glutathione (GSSG-0) at a concentration of 0.03 mM, 0.3 mM or 3 mM to make GSSG-0.03, GSSG-0.3 or GSSG-3, respectively. Paired rabbit corneas were perfused separately, and the endothelial permeability (Pac) to carboxyfluorescein was determined. Between the paired corneas perfused with GSSG-0 and GSSG-0 or GSSG-0 and GSSG-0.03, there was no significant difference in the Pac. A significant difference in this factor was seen between the paired corneas perfused with GSSG-0 and GSSG-0.3 or GSSG-0 and GSSG-3 (P less than 0.01). The ratio of GSSG-0 to GSSG-0.3 for Pac, 1.18 +/- 0.16, and that of GSSG-0 to GSSG-3, 1.14 +/- 0.07, were significantly greater than the left-right ratio for Pac obtained when the paired corneas were perfused with GSSG-0, 1.01 +/- 0.10 (mean +/- SD, n = 8) (P less than 0.025). The corneal swelling rate (micron/hr) was 7.9 +/- 4.9 for the corneas perfused with GSSG-0 and 8.4 +/- 5.4 (mean +/- SD, n = 6) for those perfused with GSSG-0.3; difference was not significant. Addition of GSSG at a concentration of 0.3 mM or more to the irrigating solution was further beneficial to the corneal endothelial barrier function and a solution containing GSSG may be safer for patients with vulnerable corneas.     

Intracellular [Na+], Na+ pathways, and fluid transport in cultured bovine corneal endothelial cells

The mechanism of fluid transport across corneal endothelium remains unclear. We examine here the relative contributions of cellular mechanisms of Na+ transport and the homeostasis of intracellular [Na+] in cultured bovine corneal endothelial cells, and the influence of ambient Na+ and HCO3- on the deturgescence of rabbit cornea. Bovine corneal endothelial cells plated on glass coverslips were incubated for 60 min with 10 microm of the fluorescent Na+ indicator SBFI precursor in HCO3- HEPES (BH) Ringer's solution. After loading, cells were placed in a perfusion chamber. Indicator fluorescence (490 nm) was determined with a Chance-Legallais time-sharing fluorometer. Its voltage output was the ratio of the emissions excited at 340 and 380 nm. For calibration, cells were treated with gramicidin D. For fluid transport measurements, rabbit corneas were mounted in a Dikstein-Maurice chamber, and stromal thickness was measured with a specular microscope. The steady-state [Na+]i in BH was 14.36+/-0.38 mM (n = mean+/-s.e.). Upon exposure to Na+ -free BH solution (choline substituted), [Na+]i decreased to 1.81+/-0.20mM (n = 19). When going from Na+ -free plus 100 microm ouabain to BH plus ouabain, [Na+]i increased to 46.17+/-2.50 (n = 6) with a half time of 1.26+/-0.04 min; if 0.1 microm phenamil plus ouabain were present, it reached only 21.78+/-1.50mm. The exponential time constants (min-1) were: 0.56+/-0.04 for the Na+ pump; 0.39+/-0.01 for the phenamil sensitive Na+ channel; and 0.17+/-0.02 for the ouabain-phenamil-insensitive pathways. In HCO3- free medium (gluconate substituted), [Na+]i was 14.03+/-0.11mM; upon changing to BH medium, it increased to 30.77+/-0.74 mm. This last [Na+]i increase was inhibited 66% by 100 microm DIDS. Using BH medium, corneal thickness remained nearly constant, increasing at a rate of only 2.9+/-0.9 microm hr-1 during 3 hr. However, stromal thickness increased drastically (swelling rate 36.1+/-2.6 microm hr-1) in corneas superfused with BH plus 100 microm ouabain. Na+ -free, HCO3- free solution and 100 microm DIDS also led to increased corneal swelling rates (17.7+/-3.6, 14.4+/-1.6 and 14.9+/-1.2 microm hr-1, respectively). The present results are explained by the presence of a DIDS-inhibitable Na+-HCO3- cotransporter and an epithelial Na+ channel, both previously found in these cells. On the other hand, the quantitative picture presented here appears a novelty. The changes we observe are consistent with pump-driven rapid exchange of intracellular Na+, and recirculation of fully 70% of the Na+ pump flux via apical Na+ channels.