Ocular rosacea. Signs, symptoms, and tear studies before and after treatment with doxycycline.

OBJECTIVE: To examine ocular signs, symptoms, and results of tear analysis in patients with cutaneous rosacea before, during, and after doxycycline therapy. DESIGN: Before-after trial. SETTING: General community. PATIENTS OR OTHER PARTICIPANTS: Thirty-nine patients with cutaneous rosacea underwent dermatologic and ocular examinations, testing of tear break-up time, and Schirmer testing at baseline and 4, 8, and 12 weeks. Six patients did not complete the study. Baseline tear break-up time and results of Schirmer test were compared with those of 13 patients without rosacea who were matched for age and sex. INTERVENTION: Patients with rosacea were given doxycycline, 100 mg daily for 12 weeks. MAIN OUTCOME MEASURE: Statistically significant (P, .05) improvement in tear break-up time. RESULT: The most frequent ocular symptoms were dryness, itching, blurred vision, and photosensitivity, all of which improved significantly with treatment. All patients had signs of ocular disease, most commonly erythema and telangiectasia, meibomian gland dysfunction, and ciliary base injection. Significant improvement (P,.05) for scales, erythema and telangiectasia, ciliary base injection, bulbar injection, papillary hypertrophy, and punctate epithelial erosions was seen. Average tear break-up time for the patients with rosacea was 5.7 seconds, which improved to 10.8 seconds after 12 weeks of treatment (P = .007). Baseline tear break-up time was significantly lower than for the comparison group of normal subjects (P = .001). There was no correlation between severity of cutaneous disease and ocular disease. CONCLUSIONS: All patients with cutaneous rosacea had some degree of ocular involvement. Tear break-up time is abnormal in patients with rosacea. Ocular erythema and telangiectasia, meibomian gland dysfunction, and short tear break-up time in patients with cutaneous rosacea are indicators of ocular rosacea. Doxycycline, 100 mg daily, will improve ocular disease and increase the tear break-up time.

Computer-simulated eye surgery. A novel teaching method for residents and practitioners.

PURPOSE: To describe an eye surgery simulator that uses a computerized graphic display to allow ophthalmic surgeons of all experience levels to enhance their surgical skills. METHODS: The eye surgery simulation environment consists of a high-speed computer graphics workstation, a stereo operating system, a wrist rest, and a position tracking stylus connected to force feedback motors. The surgeon views computer-generated images of the eye and surgical instruments through the stereo operating system and controls the position and orientation of the chosen surgical instrument by moving the stylus. During the simulated instrument-tissue interactions, three feedback motors generate component force feedback along three orthogonal axes connected by thin rigid bars to the tip of the stylus. RESULTS: The current proof-of-concept system provides a method for rapid learning experiences in a living eye simulation. Procedures can be recorded for playback and analysis, as well as for examination of techniques from different viewpoints (e.g., from inside the eye). Four simulated surgical instruments are available for use (scalpel, forceps, scissors, and phacoemulsifier). CONCLUSION: Eye surgery simulation offers both beginning and experienced ophthalmic surgeons an opportunity to learn new techniques and skills and achieve a satisfactory level of proficiency before use of that procedure in the operating room. When fully developed, this system should shorten the learning curve for new surgeons (i.e., residents) and offer an opportunity for practice before doing a difficult case or development of new techniques by experienced surgeons. The goal of replacement of current standard training methods for surgeons awaits further refinement and adjustment of the model.

Effects of calcium channel blockers on rabbit corneal endothelial function.

The effects of calcium channel antagonists and agents that alter intracellular Ca2+ mobilization on corneal endothelial function have been examined. All experiments, except where specifically designated, were performed in the continuous presence of extracellular Ca2+. Verapamil (at 50 microM) increased the swelling rate of corneas bathed in normal Ringer solution whereas nifedipine and diltiazem (both up to 100 microM) were without effect. The nifedipine analog nisoldipine caused corneal swelling at 10 microM and 50 microM but nimodipine was without effect. When briefly exposed to a Ca(2+)-free solution corneal swelling was enhanced after subsequent exposure to 50 microM verapamil in normal Ringer but not after 50 microM diltiazem in normal Ringer, indicating that Ca2+ entry from the bathing solution into the cell was important and was apparently impeded by verapamil. Cadmium (0.6 and 1 mM) but not nickel (up to 250 microM) caused swelling of corneas bathed in normal Ringer. A Ca2+ channel agonist, BAY-K-8644, alone did not influence corneal thickness but when presented to the endothelium with 50 microM verapamil the swelling rate was much reduced compared to verapamil alone. The agonist, therefore, presumably maintained some Ca2+ channels open in face of the Ca2+ channel blocker. An agent that inhibited the release of intracellular Ca2+ stores (TMB-8) caused an initial corneal swelling over the first 1.5 hr of perfusion but thereafter had no effect on corneal thickness. In the presence of continued extracellular Ca2+ one explanation for the results is that modulation of intracellular Ca2+ by agents that alter plasma membrane transfer of Ca2+ influences apical junction permeability.(ABSTRACT TRUNCATED AT 250 WORDS)

Hydrogen peroxide effects on ionic and non-ionic permeability of the rabbit corneal endothelium.

Perfusion of the isolated rabbit corneal endothelium with 0.3 mM hydrogen peroxide (H2O2) caused an increased passive permeability to bicarbonate relative to control tissues. This was accompanied by a reduction in the active flux that resulted in a reduced net bicarbonate flux. Perfusion with 0.3 mM H2O2 resulted in a marked increase in the active and net flux of sodium beginning at two hours. By four hours the net sodium flux had increased by nine-fold over control values. Perfusion with 0.3 mM H2O2 resulted in a 16% and 30% increase in endothelial permeability to inulin and dextran, respectively. Suppression of catalase activity by in vivo pretreatment with intravenous 3-aminotriazole (3AT) did not result in an increased sensitivity of the corneal endothelium to 0.2 mM H2O2: both bicarbonate and sodium fluxes were normal. Inhibition of glutathione synthesis with intravitreal buthionine sulfoximine (BSO) increased the sensitivity of the corneal endothelium to 0.2 mM H2O2 only in the case of sodium flux, with a 4.8-fold increase in net sodium flux at 3 hours after initiation of perfusion. Bicarbonate fluxes were unaffected after BSO pretreatment. The data show that ionic and non-ionic fluxes are altered by H2O2, that pretreatment with 3AT has a minimal effect on ion fluxes while BSO markedly alters sodium flux without changing bicarbonate fluxes, and that sodium and bicarbonate movement are not locked in a symport.

Effects of thymoxamine on corneal endothelium.

We have investigated the effects of 0.02 and 0.2% thymoxamine hydrochloride on the isolated rabbit corneal endothelium. The corneal swelling rate, measured by specular microscopy, indicated that 0.02% thymoxamine caused a swelling rate equal to controls while a 0.2% concentration caused a significantly increased swelling rate (34.1 vs 10.3 microns/h; P less than 0.05). The data suggests that the maximum recommended intracameral concentration of thymoxamine be 0.02% in order to allow a 10-fold safety factor for the corneal endothelium.

Corneal endothelial permeability after anterior chamber silicone oil.

One of six silicone oils, differing in both viscosity and manufacture, was infused into the anterior chambers of rabbit eyes. Polydimethylsiloxane oil, 5000 cps, caused an increased corneal endothelial permeability to inulin and dextran at 24, 96, and 168 hours after placement into the eye. Intraocular pressures were slightly elevated in the experimental eyes, compared with contralateral controls, at 24 and 144 hours after infusion. The effects of five other oils on corneal endothelial permeability were examined 168 hours after infusion. All oils increased permeability and caused thinning of endothelial cells, together with the appearance of a retrocorneal membrane, except Dow Corning Medical Fluid 360. The results indicated that contact of most silicone oils with corneal endothelium rapidly induces physiologic and morphologic changes.

Effect of PO2 and metabolic inhibitors on ionic fluxes across the isolated rabbit corneal endothelium.

Bicarbonate and sodium fluxes were measured across the isolated rabbit corneal endothelium under the influence of several inhibitors. Depression of PO2 in the bathing medium decreased net sodium movement but increased bicarbonate movement. Furosemide did not alter bicarbonate fluxes at either 10(-4) or 10(-5) M, but increased passive sodium flux leading to a decrease in net flux. Thiocyanate, at 5 x 10(-3) or 5 x 10(-2) M, decreased active bicarbonate flux and hence net flux, but had no effect on sodium fluxes. Dinitrophenol increased only the passive bicarbonate flux while decreasing both active and passive sodium fluxes, albeit unequally, leading to a decreased net flux. Ethacrynic acid affected only passive bicarbonate flux, while decreasing net sodium flux. The stilbene derivatives, SITS and DIDS caused opposite effects on both sodium and bicarbonate fluxes. SITS decreased net bicarbonate flux by decreasing active and increasing passive flux, yet increased net sodium flux. DIDS, however, increased net bicarbonate flux but decreased net sodium flux. The results may be explained by current models for endothelial ion transport that include a Na+/H+ antiport and a HCO3-/Na+ symport system in parallel with an independent pathway for HCO3- exit from the endothelial cells. When compared with prior corneal swelling data using these same inhibitors, the maintenance of corneal thickness appears to be dependent on the variation of ion fluxes from normal values, and the dissociation of the two active ion fluxes. In addition, there appears to be a significant ability of ion transport systems to compensate for disturbances to other ion exchange or transport mechanisms.

Morphologic and physiologic effects of thymoxamine on corneal endothelium and ciliary processes.

Either 0.2 or 0.02% thymoxamine hydrochloride was perfused across the endothelium of isolated rabbit corneas in a specular microscope. Treated corneas did not swell, compared to controls, when perfused with 0.02% thymoxamine, but swelled at 34 microns/hr after 0.2% thymoxamine compared to 10 microns/hr of controls. Morphologically, 0.2% thymoxamine caused many intracellular vacuoles, with far fewer changes after 0.02%. Replacement of the aqueous humor with 0.02% thymoxamine caused no effect on either the corneal endothelium or ciliary process, whereas 0.2% caused small, but significant, changes in both tissues.

Effects of silicone oils on corneal endothelial permeability.

Silicone oils, varying by viscosity and manufacturer, were infused into rabbit anterior chambers. Polydimethyl-siloxane oil, 5000 cps, increased corneal endothelial permeability to inulin (mw 5000) and dextran (mw 60000) when measured in vitro at 1, 4 and 7 days after ocular infusion. The effects of five other oils were measured at 7 days after infusion. Four of the oils increased endothelial permeability and induced similar morphological changes. Dow Corning Medical Fluid 360 had no effect on either permeability or morphology of the endothelium. These results show that contact of most silicone oils with corneal endothelium rapidly induces physiological and morphological changes. If these oils, when used as a retinal tamponade, gain access to the cornea they should be removed quickly to avoid the rapid initiation of physiologic changes.

Oxygen free radicals and corneal endothelium.

We have demonstrated that corneal endothelial cells are susceptible to damage from hydrogen peroxide. Hydrogen peroxide perfusion of corneal endothelium results in physiologic and anatomic disruption of corneal endothelial cells with resulting swelling of the corneal stroma. Enzymatically generated superoxide anion produced endothelial damage which could not be blocked with superoxide dismutase (which removes superoxide anion from the system), but could it be blocked with catalase which removes hydrogen peroxide from the system. This indicated that the superoxide anion was not the toxic agent, but rather hydrogen peroxide, its dismutation product. Enzymatically generated oxygen free radical in the anterior chamber of the rabbit eye resulted in an increased iris vascular permeability which persisted for up to 24 hours following anterior chamber injection of hypoxanthine and xanthine oxidase. Young animals were better able to withstand oxidative stress to the eye then were older animals. The physiologic effects of oxidative stress in young animals can be made comparable to that in the older animals by depletion of catalase with 3-aminotriazole.

Corneal storage in MK medium and K-Sol. Effect on ionic and non-ionic fluxes.

Rabbit corneas were stored at 4 degrees C for 3, 7 or 14 days in either modified MK medium or K-Sol. Corneal endothelial permeability to inulin following storage in modified MK was significantly less at each time examined than that found in corneas stored for either 3, 7 or 14 days in K-Sol. Inulin permeability after storage in K-Sol was increased at all times relative to unstored control corneal tissue, but only at 7 and 14 days in MK medium. Dextran permeability was similar following 3 days of storage in either solution, but dextran permeability following storage in modified MK was significantly less than the values found in corneas stored for 7 and 14 days in K-Sol. Dextran permeability was not significantly increased relative to control, at any storage time in MK medium but was increased at 7 and 14 days in K-Sol. Inulin and dextran permeabilities after storage in MK medium were maintained more closely to values found in fresh tissue than corneas stored in K-Sol. Net endothelial sodium fluxes following storage in modified MK medium were markedly less than those found in corneas stored for 3, 7 and 14 days in K-Sol. Net sodium fluxes are maintained better in K-Sol than in MK medium relative to control values. Net bicarbonate fluxes following storage in modified MK medium were significantly less than the 3-day values in K-Sol, but similar to the values after 7 and 14 days of K-Sol storage.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of carbachol on rabbit corneal endothelium.

We investigated the current reformulation of the commercially available intraocular preparation of carbachol for its effect on corneal thickness and endothelial cell ultrastructure. A 15-minute perfusion in the specular microscope with 0.01% carbachol caused a significant increase in the corneal swelling rate, which returned to control values after 75 minutes of reperfusion with Krebs Ringer bicarbonate solution. Continued reperfusion with Krebs Ringer bicarbonate solution led to the final average corneal swelling rate being statistically similar to that of controls over the entire 180-minute perfusion period. Transmission and scanning electron microscopy demonstrated no changes in endothelial cell ultrastructure.

Rabbit corneal endothelial physiologic and morphologic characteristics following storage in MK medium and K-Sol.

Rabbit corneas were stored for 3, 7, 10 or 14 days in MK medium or K-Sol. Corneas were thinner immediately following removal from K-Sol than following removal from MK medium at all time periods studied. Following 3 days of storage, corneas stored in both solutions showed similar temperature reversal characteristics. Corneas stored for 7 and 10 days in MK medium also temperature reversed, whereas those stored in K-Sol did not. Corneas stored for 14 days in both solutions swelled following mounting in the specular microscope. Endothelial cell morphology was similar following storage in MK medium and K-Sol at all time periods studied. It is postulated that a persistant osmotic gradient is present across the endothelium following storage in K-Sol. This osmotic gradient occurs because of retention of chondroitin sulphate in the corneal stroma thereby preventing early temperature reversal.

Effect of ambient pH on corneal endothelial sodium fluxes.

Sodium fluxes across the isolated rabbit corneal endothelium were measured as a function of ambient pH and bicarbonate concentrations. At pH 7, the net sodium flux (J endo str net) was effectively obliterated at all bicarbonate concentrations, relative to controls at pH 7.5. At pH 8, the net sodium flux was obliterated at 5 mM bicarbonate, was unchanged at 25 mM bicarbonate, and was enhanced at 40 mM bicarbonate, relative to controls at pH 7.5. The decrease in net sodium flux under these conditions relative to previously demonstrated maintenance of bicarbonate fluxes under almost identical conditions suggest that sodium and bicarbonate transverse the endothelium by different routes. In addition, changes in unidirectional fluxes, when ambient pH is decreased from 7.5 to 7, are not equal, illustrating that the fluxes probably occur via different pathways. A complex interrelationship exists between sodium, bicarbonate, and proton movement.

Effect of dextran-bound acetazolamide on rabbit corneal endothelial ion fluxes.

The effects of a carbonic anhydrase inhibitor, acetazolamide, bound to a 72,000 dalton dextran (DBI) on bicarbonate and sodium fluxes across the isolated rabbit corneal endothelium have been examined. When DBI was present on the aqueous-facing endothelial surface, there was a marked inhibition of both the stromal to endothelial unidirectional and net flux of bicarbonate. This suggests that the exit of bicarbonate from the cell into the aqueous-facing solution is influenced by the enzyme carbonic anhydrase. No change was found in sodium fluxes under these same incubation conditions. When DBI was present on the stromal-facing surface of the endothelium, no changes were found in unidirectional or net bicarbonate fluxes; the sodium flux from stroma to endothelium was increased, however, with no change in net flux. This data implies that the link between sodium and bicarbonate movement across the endothelium is not a direct coupling between the transport of the two ions in the form of a symport (Na+:HCO3-) at the apical cell border.

Lupus erythematosus keratoconjunctivitis.

A patient with systemic lupus erythematosus and a three-month history of bilateral follicular conjunctivitis had both a superficial and deep culture-negative keratitis. Although SLE interstitial keratitis is uncommon, it must be recognized as a definite clinical entity. Its prompt recognition and treatment with corticosteroids may quickly bring about improvement, whereas if it is unrecognized, corneal scarring and loss of vision may follow.

BSS and BSS-plus: effect on corneal endothelial ionic and non-ionic fluxes.

Rabbit corneas were mounted in water jacketed chambers and the endothelial surface perfused with either BSS (Balanced Salt Solution) or BSS-Plus for 3 hr. Unidirectional and net fluxes of sodium were similar in both groups of corneas. Bicarbonate fluxes in BSS-Plus were statistically similar to those in Krebs-Ringer solution. Bicarbonate fluxes could not be determined with BSS because the solution does not contain bicarbonate. In addition, there was no statistically significant alteration of inulin or dextran permeability when comparing perfusion with BSS and BSS-Plus. From this study it appears that BSS and BSS-Plus are comparable in their ability to maintain corneal endothelial physiologic function during in-vitro perfusion. This is in contrast to previous work which showed that BSS-Plus induced less endothelial morphologic change than BSS. It is concluded that morphologic alterations may be more sensitive parameters of endothelial stress than are fluxes and permeabilities.