Vitrectomy in endophthalmitis. Results of study using vitrectomy, intraocular antibiotics, or a combination of both.

A rabbit model of endophthalmitis was produced by inoculating Staphylococcus epidermidis and S aureus into the vitreous cavity. Elimination of microorganisms was compared using intravitreal administration of 0.1 mg of gentamicin alone, vitrectomy alone, and a combination of gentamicin and vitrectomy. In the case of S. epidermidis treated 24 hours after inoculation, all untreated eyes were culture-positive at one week, eyes treated with gentamicin alone or intravitreal gentamicin in combination with vitrectomy were all culture-negative, and vitrectomy alone rendered half of the eyes culture-negative. In the case of S aureus, eyes not treated by vitrectomy alone were all culture-positive at one week. Eyes treated with intravitreal gentamicin 25 to 31 hours after infection were culture-negative for S aireus in 33% at one week, while eyes treated with combined vitrectomy plus intraocular gentamicin were culture-negative in 83% of cases. When treatment was delayed 40 to 49 hours after inoculation of S aureus, intravitreal rendered 50% culture-negative at one week, while vitrectomy combined with intravitreal gentamicin eliminated the infection in 100% of eyes.

The diagnosis and management of keratomycoses. I. Cause and diagnosis.

Causative isolates, clinical features, and laboratory studies are reported for sixty-one cases of culture-proved mycotic keratitis. Isolates are categorized into four groups, including 36 Fusarium solani, 11 other Moniliaceae species, seven Dematiaceae, and seven yeasts. Of the 61 patients, 42 were men. Mild outdoor trauma was sustained in 14 of 24 cases. Patients were often referred with a clinical diagnosis of presumed fungal keratitis, within one week of symptom development, and usually had not received topically applied steroids prior to referral. Laboratory diagnosis necessitates prompt corneal scrapings, preferably stained with Giemsa or Gram, and culture on Sabouraud and blood agar maintained at room temperature, with growth usually evident by 48 hours.

Papanicolaou cytology in the diagnosis and management of external ocular tumors.

Papanicolaou smear cytology is a simple diagnostic procedure for the evaluation and management of external ocular lesions. A variety of benign and neoplastic lesions are examined for correlation of clinical, cytologic, and histologic impressions. Accurate correlation in 38 of 41 eyes substantiates the reliability of Papanicolaou smear cytology to differentiate benign from neoplastic ocular lesions. Cytology proved useful in the preoperative evaluation of squamous cell carcinoma and related tumors of the conjunctiva, and malignant melanoma of the conjunctiva. Cytology also was valuable in the detection of tumor recurrence in squamous cell carcinoma and carcinoma-in-situ of the conjunctiva, and in Meibomian gland carcinoma of the eyelid. The accessibility of ocular tumors makes them particularly well suited for cytologic examination.

Control of corneal astigmatism following cataract extraction by selective suture cutting.

We evaluated the ability of selective suture cutting to reduce postoperative corneal astigmatism in 75 eyes of 68 patients who underwent extracapsular cataract extraction with posterior chamber intraocular lens implantation. Keratometric and refractive measurements were obtained before and at selected intervals (3, 6, 10, 26, and 52 weeks) after surgery. The number of sutures cut during the sixth week after surgery was based on the degree of astigmatism (0.00 to 2.00 diopters (D), no sutures cut; 2.25 to 3.00 D, one suture cut; 3.25 to 4.00 D, two sutures cut; greater than or equal to 4.25 D, three sutures cut). Our analysis demonstrated the following: (1) a spontaneous reduction of 0.5 D in surgically induced astigmatism in eyes without suture cutting, (2) an additional reduction of 1.2 D in postoperative astigmatism for each suture cut, and (3) attainment of 75% to 93% of the total effect of suture cutting within four weeks. The final astigmatism one year after surgery had increased by a mean of 0.9 D, exhibited predominantly with-the-rule properties, and showed no significant difference among the four groups of patients. Vector analysis revealed that only small shifts in the axis of astigmatism occurred after suture cutting. A strong correlation between the keratometric and subjective refractive measurements during all postoperative examinations indicated that corneal astigmatism is primarily responsible for postoperative astigmatism.

Comparison of MacKay-Marg, Goldmann, and Perkins tonometers in abnormal corneas.

Intraocular pressure was determined manometrically in owl monkey eyes with normal and edematous corneas. The pressure was measured with Perkins hand-held, Goldmann, and MacKay-Marg tonometers to compare relative accuracy. At monometric levels of 20, 30, and 40 mm Hg, a statistically significant difference in readings was found between normal and edematous corneas with use of the Perkins and Goldmann applanators, whereas no such difference was found with the MacKay-Marg tonometer. The MacKay-Marg tonometer most closely approximated the true intraocular pressure in edematous corneas.

The diagnosis and management of keratomycoses. II. Medical and surgical management.

Medical management of 61 cases of mycotic keratitis, including the use of natamycin (pimaricine) in 53 cases, resulted in successful healing in 46 cases. A final visual acuity of 20/40 or better was achieved in 25 cases by medical therapy alone. Thirteen cases were considered medical treatment failures, and 11 necessitated therapeutic surgery. A final visual acuity of 20/70 or better was achieved in six of these cases, including five of nine therapeutic penetrating keratoplasties. In four cases, ulceration had progressed despite natamycin treatment, but fungal cultures were negative at the time of therapeutic surgery.

Clinical specular microscopy and intraocular surgery.

Seventy patients were studied prospectively, using the clinical specular microscope, to compare and evaluate the effects on the corneal endothelium of routine intracapsular cataract extraction (ICCE), ICCE with intraocular lens (IOL) implantation, phacoemulsification, and penetrating keratoplasty. Results showed a 6% cell loss after ICCE, 12% loss after IOL insertion, 14% after phacoemulsification, and 21% loss after penetrating keratoplasty. Both phakic and aphakic grafts had 21% cell loss. Within the aphakic keratoplasty group, there was a 10% cell loss in patients who maintained an intact vitreous face compared with a 26% cell loss in patients who underwent an anterior vitrectomy (P less than .1).

Determinants of graft clarity in penetrating kerotoplasty.

A total of 240 consecutive penetrating keratoplasties with a minimum six months' follow-up were examined to determine the influence of preoperative prognosis, donor age, death-enucleation time, death-surgery time, and surgical experience on graft clarity. Graft results, classified as clear, nebulous, or edematous, were compared within two prognostic categories in aphakic, combined, and phakic keratoplasty groups. Preoperative prognostic groups and surgical experience appeared to be most important in affecting graft clarity, whereas the donor age, death-enucleation time, and death-surgery time seemed to have little effect.

Diagnostic limulus lysate assay for endophthalmitis and keratitis.

The limulus lysate assay is an inexpensive, reliable, and rapid means of detecting and presence of Gram-negative endotoxin. In all ten cases of experimentally induced Proteus endophthalmitis in rabbits, the assay was positive, and the assay was appropriately negative in all ten cases of Staphylococcal endophthalmitis, ten cases of Candida endophthalmitis, and ten cases of sterile endophthalmitis in rabbits. In a clinical assessment of keratitis, the assay of corneal scrapings was positive in 11 of 13 Gram-negative corneal ulcers. In a similar study of clinical endophthalmitis, both Gram-negative cases had a negative limulus assay, but two cases are insufficient to be conclusive. The assay may prove to be a useful adjunct both to standard diagnostic evaluations and in the rapid direction of appropriate therapy for these conditions.

Chronic conjunctivitis associated with 'floppy eyelids'.

The floppy eyelid syndrome represents a chronic conjunctivitis characterized by a triad of (1) a loose upper lid that readily everts on elevating the lid, (2) a soft rubbery tarsus that can be folded on itself, and (3) a diffuse papillary conjunctivitis. Seven cases of this condition are reported, all involving men, and four involving obese patients.

Endogenous Aspergillus endophthalmitis in drug abusers.

One of two cases of endogenous Aspergillus endophthalmitis in abusers of intravenously administered drugs was treated successfully by subtotal pars plana vitrectomy; amphotericin B administered by intravitreal, periocular, and systemic routes; and flucytosine administered systemically. Aspergillus sp should be considered a possible pathogenic organism in drug abusers with endogenous endophthalmitis. An aggressive diagnostic and therapeutic approach may result in preservation of useful vision.

Neodymium-YAG laser posterior capsulotomy. Central corneal endothelial cell density.

We determined central corneal endothelial cell densities in 39 eyes in a masked fashion before and after Q-switched neodymium-YAG laser posterior capsulotomy. The mean preoperative endothelial cell count was 1,840 cells per square millimeter (SD, 611 cells per square millimeter; range, 522 to 2,742 cells per square millimeter). The mean postoperative endothelial cell count was 1,798 cells per square millimeter (SD, 600 cells per square millimeter; range, 478 to 2,750 cells per square millimeter). The difference between the mean preoperative and postoperative cell density was 42 cells per square millimeter (2.3% cell loss). There was no significant correlation between central corneal endothelial cell loss and the laser energy used, pseudophakic status (posterior chamber intraocular lens vs no intraocular lens), size of capsulotomy, vitreocorneal touch, or preoperative corneal endothelial cell density.

Laser iridotomy for aphakic pupillary block.

In six patients with aphakic pupillary block, argon laser produced full-thickness iridotomies that succeeded in restoring anterior chamber and intraocular pressure to normal. In one case, it was shown that location of iridotomy is important in determining whether or not iridotomy will succeed in relieving aphakic pupillary block. In two cases, pupillary block had resulted in shallowing of the anterior chamber without pressure elevation. Cases reported here show that continuous laser energy, such as from an argon laser, can produce a full-thickness opening, in contrast to the short duration pulse of the ruby, which will destroy only pigmented structures without producing a full-thickness iris opening. This need to produce some degree of spread of damage to unpigmented tissue elements in order to achieve and iris opening is a problem in phakic patients where injury to the lens must be avoided.

Toxicity, efficacy, and clearance of intravitreally injected of cefazolin.

The intravitreal injection of cefazolin sodium was evaluated with respect to its potential toxicity, efficacy, and clearance in the treatment of bacterial endophthalmitis in the rabbit model. Toxicity evaluation included indirect ophthalmoscopy, scotopic electroretinogram, and light microscopy of histopathologic specimens. Rabbits given doses up to and including 2.25 mg of cefazolin sodium showed no evidence of toxic reaction by these criteria after two weeks of observation. When injected within 24 hours of inoculation, 2.25 mg of cefazolin sodium sterilized five of six rabbit eyes infected with Staphylococcus aureus. Following intravitreal injection of 2.25 mg of cefazolin sodium, vitreous concentrations in excess of the minimum inhibitory concentration for S aureus were maintained for five days. A 2.25-mg dose of cefazolin sodium appears to be nontoxic, effective, and biologically active against susceptible organisms during a prolonged period, when injected into the vitreous cavity of the pigmented rabbit in vivo.

Zonular dialysis during extracapsular cataract extraction in pseudoexfoliation syndrome.

Five patients with pseudoexfoliation syndrome (PES) and glaucoma developed extensive zonular dialyses during extracapsular cataract extractions. Weakness of the lens zonules or their attachments to the ciliary processes, which has been described in association with PES, may explain this complication. We believe that patients with PES are at particular risk for developing large zonular dialyses during extracapsular surgery. Preoperative phakodonesis, anterior chamber depth asymmetry, and excessive lens movement during the anterior capsulotomy should alert the surgeon to this problem.

Intravitreal vancomycin. Retinal toxicity, clearance, and interaction with gentamicin.

Some of the gram-positive isolates from exogenous bacterial endophthalmitis cases treated at our institution have been found to be resistant to either cefazolin sodium, gentamicin sulfate, or both. However, all of these isolates have been sensitive to vancomycin. These findings prompted us to reevaluate the retinal toxicity and clearance of intravitreal vancomycin in pigmented rabbits. Doses up to 2 mg were found to be nontoxic in phakic and aphakic-vitrectomized eyes. Clearance was determined in phakic and aphakic-vitrectomized rabbit eyes with or without intact lens capsules. The antibiotic was cleared most slowly in phakic eyes. Aphakic-vitrectomized eyes without an intact lens capsule cleared antibiotic most rapidly, while aphakic-vitrectomized eyes with intact capsules exhibited an intermediate clearance rate. In addition, the interaction between vancomycin and gentamicin on gram-positive endophthalmitis isolates was found to be additive or synergistic depending on the bacterial species. Based on these data, we recommend the combination of vancomycin and an aminoglycoside as the initial antibiotic therapy for exogenous bacterial endophthalmitis.

Late development of ulcerative keratitis in radial keratotomy scars.

We describe three patients in whom ulcerative keratitis developed seven months to 2 1/2 years after uncomplicated radial keratotomy. All ulcers occurred along keratotomy scars within the interpalpebral fissure or inferiorly. At least one case developed spontaneously; another patient may have sustained an unrecognized corneal injury. The third patient was using an extended-wear soft contact lens for correction of residual myopia. Bacteria were isolated from two of the ulcers. With therapy, all ulcers healed without reducing visual acuity. We hypothesize that the altered structure of the slowly healing keratotomy incisions coupled with redistribution of the tear film due to changes in corneal topography may result in intermittent epithelial irregularities. Alone or in conjunction with any form of mechanical or hypoxic corneal injury, these changes along keratotomy incisions may predispose to corneal infection.

Amelioration of experimental lipid keratopathy by photochemically induced thrombosis of feeder vessels.

The photochemical interaction between intravenously injected rose bengal dye and 514.5-nm argon laser irradiation was employed to initiate permanent thrombotic occlusions in the corneal neovasculature of rabbit eyes with experimentally induced lipid keratopathy. This photothrombotic procedure did not produce corneal edema or polymorphonuclear leukocyte infiltration, as was found in previous studies employing laser-induced photocoagulation. Thus, by means of the new technique, we avoided the formation of additional neovascularization and lipid deposition. Vascular occlusion by photothrombosis, effected with 8.5 times less incident intensity and 27.5 times less total light exposure than with photocoagulation, yielded an average reduction of corneal cholesterol content of 36% as opposed to an increase of 24% found with previous argon laser photocoagulation.

Gonococcal keratoconjunctivitis.

Gonococcal keratoconjunctivitis is a potentially devastating infection, because Neisseria gonorrhoeae can cause a rapid, severe, ulcerative keratitis resulting in visual loss. The therapeutic decision making process is complicated by the necessity for prompt, effective parenteral therapy, frequent coinfection with other sexually transmitted diseases, and emergence of antibiotic resistance. Because of the evolving problem of antibiotic resistance and the need for cost containment, the current recommendations of hospitalization for intravenous penicillin may need to be modified. The third generation cephalosporin, ceftriaxone, has properties that suggest it may be the best available antimicrobial agent as a single-dose treatment of gonococcal conjunctivitis. Spectinomycin may be a useful alternative in the penicillin-allergic adult patient.

Animal model of Fusarium solani keratitis.

Because of the frequency of fungal keratitis due to Fusarium solani, we needed a sustained, progressive infection in an animal model to determine the mechanisms of pathogenicity and to evaluate the new antifungal agents. Pigmented rabbits interlamellarly injected with actively germinating conidia from lyophilized temperature-tolerant strains of F. solani produced sustained culture-positive ulcers in high percentage of eyes at two and three weeks, pretreatment with subconjunctival corticosteroids was necessary. Histopathology, although a poor index of infectivity since some corneas with plentiful hyphal fragments had negative cultures, simulated human fungal pathology.