Endophthalmitis caused by gram-negative organisms.

The medical records of 52 patients (53 eyes) with culture-proven gram-negative endophthalmitis between January 1982 and December 1990 were reviewed. Pseudomonas aeruginosa (23% [12/53]) and Haemophilus influenzae (19% [10/53]) were the most frequent isolates in this series. Overall, 26 (49%) of 53 treated patients achieved 20/400 or better visual acuity. Fifty-two (98%) of the original 53 gram-negative isolates were sensitive to the aminoglycoside antibiotics. To determine their sensitivity to recently developed antibiotics, 35 of the isolates were again grown on culture media and their sensitivities to ceftazidime, ciprofloxacin, and imipenem were obtained. Only ceftazidime demonstrated in vitro efficacy for all the organisms tested.

Retinal toxicity of human tissue plasminogen activator in vitrectomized rabbit eyes.

The retinal toxicity of human tissue plasminogen activator in normal rabbit eyes has recently been reported. We now report the retinal toxicity of tissue plasminogen activator in three groups of vitrectomized rabbit eyes. Group 1 underwent gas compression of the vitreous followed by tissue plasminogen activator injection in doses of 25, 50, and 100 micrograms (all doses were administered in 100 microL of fluid). Group 2 underwent lensectomy and vitrectomy followed by tissue plasminogen activator injection of 100 micrograms. Group 3 underwent lensectomy, vitrectomy, and complete fluid/gas exchange prior to injections of 12.5 and 25 micrograms of tissue plasminogen activator. Control eyes received 100 microL of balanced salt solution. In group 1, no retinal toxic reactions were observed after administration of 25 or 50 micrograms of tissue plasminogen activator, but all eyes receiving 100 micrograms demonstrated retinal damage on ophthalmoscopy, electroretinography, and light microscopy. In group 2, no retinal toxic reactions were seen after administration of 100 micrograms of tissue plasminogen activator. In group 3, two of 11 eyes receiving 25 micrograms of tissue plasminogen activator demonstrated toxic retinal changes by ophthalmoscopy, electroretinography, and light microscopy. These results suggest that gas compression of the vitreous does not significantly alter the toxic changes seen caused by tissue plasminogen activator. While lensectomy and vitrectomy appears to widen the therapeutic window for tissue plasminogen activator, the margin of safety is reduced with the addition of a large gas bubble.

Retinal toxicity of recombinant tissue plasminogen activator in the rabbit.

We studied the dose-dependent retinal toxicity of the available commercial preparation of human recombinant tissue plasminogen activator in the normal rabbit eye. Tissue plasminogen activator was injected into the midvitreous cavity of albino rabbits in doses (per 100 microL) of 25, 50, 75, and 100 micrograms. Control eyes received 100 microL of balanced salt solution or tissue plasminogen activator vehicle. No evidence of a retinal toxic reaction was seen in eyes receiving 25 micrograms of tissue plasminogen activator. One of four eyes injected with 50 micrograms showed loss of photoreceptor cells by light microscopy. Severe retinal damage was seen by ophthalmoscopy, electroretinography, and light microscopy in three of four eyes receiving 75 micrograms of tissue plasminogen activator and in all eyes treated with 100 micrograms of tissue plasminogen activator or equivalent vehicle. These results suggest that the commercial recombinant tissue plasminogen activator formulation has a narrow margin of safety in nonvitrectomized eyes and that a component of the vehicle is the toxic factor.

Retained lens fragments after phacoemulsification manifesting as marked intraocular inflammation with hypopyon.

We reviewed the medical records of four patients with marked intraocular inflammation and hypopyon as the initial manifestation of retained lens fragments after phacoemulsification. The severe inflammatory reaction occurred between one month and one year after the cataract extraction. All four patients underwent pars plana vitrectomy to remove the lens fragments. The vitreous specimens were cultured to rule out infectious endophthalmitis. In all patients, no organisms were isolated from the vitrectomy specimens placed on both aerobic and anaerobic media. All patients had improved vision and resolution of the marked intraocular inflammation after vitrectomy. Echography was useful in establishing the diagnosis in these uncommon cases.

Vitrectomy for complications of proliferative diabetic retinopathy. Functional outcomes.

PURPOSE: To quantitate the effect of diabetic vitrectomy on each patient's visual system and function. METHODS: The anatomic and visual acuity outcomes of diabetic vitrectomy among three surgeons at one institution over a 4-year time interval were studied. Outcome parameters describing the visual system function were defined and applied to quantitate the effect on visual system disability. RESULTS: The anatomic and visual success rates and prognostic factors in the cohort of 213 patients by standard statistical analysis were similar to previous studies. The study eye was the eye with better vision in 68 (32%) patients at the conclusion of the study, and vision was equal to the fellow eye in 35 (16%) patients. The mean impairment of the visual system by the American Medical Association Guidelines for Disability was reduced from 61% preoperatively to 50% postoperatively. The factor associated with greatest degree of reduction of impairment of the visual system was a previtrectomy diagnosis of vitreous hemorrhage. There was at least a one-step improvement of visual system function in 65 (31%) patients. CONCLUSIONS: Vitrectomy for complications of severe proliferative diabetic retinopathy is especially valuable in improving the patient's overall visual function.