Evaluation of intraocular reactivity to metallic and ethylene oxide contaminants of medical devices in a rabbit model.

OBJECTIVE: To evaluate the intraocular reactivity to metallic and ethylene oxide (EO) contaminants of ophthalmic devices in rabbits. DESIGN: Two experimental animal studies. PARTICIPANTS: Thirty-five New Zealand white rabbits. METHODS: A metallic exposure study and an EO exposure study were performed. In the first study, both eyes of 25 rabbits were equally allocated to intracameral injections of alumina 0.2 µg, alumina 20 µg, copper sulfate 0.4 µg, copper sulfate 20 µg, or an aqueous control. In the second study, 10 rabbits were allocated (5 per group) to receive intracamerally an ophthalmic viscosurgical device (OVD) exposed to EO or not exposed to EO (control). All eyes were examined by slit lamp at baseline and 3, 6, 9, 24, 48, and 72 hours after exposure, with dilated indirect ophthalmoscopy being performed at 24 and 72 hours. Tonometry was performed only in the first study. MAIN OUTCOME MEASURES: Grade of corneal clouding, anterior chamber (AC) flare, AC cells, AC fibrin, iridal hyperemia, cell and fibrin on the lens surface, vitreous haze and cells, lens opacities, intraocular pressure, and onset time. RESULTS: For metallic compounds at the study's low doses, mean inflammatory grades were 0.2 or less above the control for all responses at all time points. For the high-dose alumina, mean inflammatory grades peaked at 6 to 9 hours at 0.5 to 0.7 above the control responses for conjunctival congestion, iris hyperemia, AC cells, flare, and fibrin and declined over the remaining time points. For the high-dose copper sulfate, mean inflammatory grades peaked between 3 and 24 hours at 1.2 to 1.8 above the control responses for conjunctival congestion, iris hyperemia, AC cells, flare, fibrin, and corneal clouding, then subsequently declined. The intraocular pressure changes appeared significant for only high-dose copper sulfate, with mean declines of 4.3 to 7.5 mmHg at 6 to 72 hours. No clinically meaningful differences in ocular inflammation were observed between the OVD exposed to EO and the OVD not exposed to EO. CONCLUSIONS: Alumina and copper sulfate did not cause clinically meaningful ocular inflammation at the low study levels (levels expected with ophthalmic devices). Ethylene oxide exposure of an OVD was not associated with inflammation.

An investigation of enzymatic detergents as a potential cause of toxic anterior segment syndrome.

OBJECTIVE: To investigate whether enzymatic detergents used in cleaning ophthalmic surgical instruments can cause toxic anterior segment syndrome (TASS)-like responses in a rabbit model. DESIGN: Randomized, investigator-masked, controlled experimental animal study. PARTICIPANTS: Thirty-five New Zealand white rabbits. METHODS: The rabbit eyes were randomized into 7 treatment groups to receive intracameral injection of 1 of 3 different doses of Medline Dual Detergent or Enzol Detergent, or sterile limulus amoebocyte lysate reagent water as a control. The eyes were evaluated for anterior segment inflammation at baseline and at 1, 3, 6, 24, 48, and 72 hours after treatment by slit-lamp biomicroscopy. MAIN OUTCOME MEASURES: Anterior chamber (AC) inflammation, including cells, flare, fibrin, and iris injection; time course of inflammation; and residual detergent levels in luminated instruments. RESULTS: Moderate to marked injection of the iris vessels was seen as early as 1 hour after treatment with the enzymatic detergents in 41 of 60 eyes, with the response being more severe in the Enzol Detergent-exposed eyes. Severe iris hemorrhages were accompanied by blood in the AC in 13 eyes, which usually persisted through 72 hours, with an associated increase in AC cell and flare. Corneal haze was present in 52 of 56 eyes 1 hour after treatment, but was mild and resolved within 24 hours in all but the Enzol 4.5%-exposed eyes. Median AC cell and flare peaked at 6 hours and resolved by 48 hours. CONCLUSIONS: Enzymatic detergents caused a severe but unusual response from the iris when injected intracamerally into rabbit eyes. This response has not been reported in humans with TASS. The time course of inflammation was faster (peak at 6 hours) and resolved more quickly (within 48 hours) than TASS. Simulated cleaning and extraction studies indicate that the level of residual detergent to which a patient could be exposed is significantly less than the lowest dose used in this study. Because that low dose caused no significant observations other than injection of the iris vessels, these results do not support residual enzymatic detergents on surgical instruments as a cause for TASS.

The Food and Drug Administration's Proactive toxic anterior segment syndrome Program.

Toxic anterior segment syndrome (TASS) is a rare inflammatory condition usually observed within the first 48 hours after uncomplicated anterior segment surgery. Over the decades since its initial description, a number of TASS outbreaks have been reported. For a few of these outbreaks, the inciting factors were identified, but for the majority, the precipitating factors were often postulated but not confirmed. In light of the limitations identified in these outbreak investigations, the Food and Drug Administration's (FDA's) Center for Devices and Radiological Health staff has embarked on a number of activities aimed at mitigating medical device-related TASS outbreaks. Under the FDA-designed Proactive TASS Program (PTP), FDA scientists have conducted animal studies to better explore the inflammatory potential of suspected ophthalmic device contaminants implicated in prior cases of TASS. For contaminants displaying a TASS-like reaction in these animal models, the FDA scientists have developed analytic test methods to measure the level of those contaminants in or on ophthalmic devices. Moreover, FDA researchers have developed methods to better capture the clinical information necessary to assist investigations of potential future outbreaks. Last, the FDA has partnered with the Centers for Disease Control and Prevention to facilitate a potential TASS investigation, including expediting the analysis of potentially contaminated medical devices. The PTP is an example of the FDA proactively developing test methods and disease surveillance methods geared toward protecting the public's health.

Rabbit intraocular reactivity to endotoxin measured by slit-lamp biomicroscopy and laser flare photometry.

OBJECTIVE: To evaluate the ocular reactivity of the rabbit to an intracameral injection of a dispersive ophthalmic viscosurgical device (OVD) containing various levels of bacterial endotoxin using slit-lamp biomicroscopy and laser flare photometry. DESIGN: Experimental, randomized, masked animal study. PARTICIPANTS: Thirty Dutch-Belted rabbits. METHODS: The rabbits were randomized into 6 groups to receive 0.05 ml of a hydroxypropyl methylcellulose-based dispersive OVD to which had been added one of 5 different doses of bacterial endotoxin ranging from 0.02 to 1.4 endotoxin units (EUs) or a vehicle control to both eyes. The eyes were evaluated for anterior segment inflammation at baseline and 3, 6, 9, 24, 48, and 72 hours after injection using slit-lamp biomicroscopy and laser flare photometry. MAIN OUTCOME MEASURES: Corneal clarity and anterior chamber (AC) inflammation. RESULTS: All the corneas remained clear throughout the study. Anterior chamber cells were seen at 6, 9, and 24 hours in 60% to 100% of the eyes intracamerally injected with endotoxin-containing OVD, and the response declined rapidly after 24 hours. A dose-response effect was seen between the concentration of endotoxin and the AC cell response. The aqueous flare response in eyes injected with the 2 highest doses of endotoxin was significantly greater (P<0.05) than that of controls. The amounts of fibrin observed in the AC were random, with no apparent dose-response effect seen. The flare values as obtained by laser flare photometry were consistent with the slit-lamp biomicroscopy flare findings up to grade 3+. However, the increase in laser flare value seemed to level off in eyes with more than 3+ flare. Neither measure of flare correlated with endotoxin level. CONCLUSIONS: Among the parameters evaluated in this study, the AC cell response, evaluated by slit-lamp biomicroscopy and graded using a standard grading system, was found to be the most reliable indicator of the amount of endotoxin in the dispersive OVD. The use of laser flare photometry alone does not seem to be useful in detecting an ocular response to endotoxin contamination in OVDs.

Evaluation of intraocular reactivity to organic contaminants of ophthalmic devices in a rabbit model.

OBJECTIVE: To evaluate the intraocular reactivity to organic contaminants of ophthalmic devices in the rabbit. DESIGN: Experimental animal study. PARTICIPANTS: Fifty New Zealand white rabbits. METHODS: The rabbits were allocated to 10 groups of 5 each to receive 2 different doses of human albumin and nonhuman nucleic acids and their respective vehicle controls, a denatured cohesive ophthalmic viscosurgical device (OVD) and a denatured dispersive OVD and their respective nondenatured controls. All 10 eyes in each treatment group received bilateral intracameral injection of the test materials. All the eyes in the study were examined by slit-lamp biomicroscopy at baseline and 6, 9, 24, 48, and 72 hours. Pachymetry was also performed on eyes exposed to albumin, protein vehicle control, and the OVDs at these time points. MAIN OUTCOME MEASURES: Corneal thickness, grade of corneal clouding, anterior chamber (AC), cells, flare and fibrin, iridal hyperemia, cell and fibrin on lens surface, and onset time. RESULTS: There were no inflammatory signs in any eyes exposed to human albumin. Anterior chamber cells (1+ to 3+) and flare and fibrin (1+ to 2+), along with cells and fibrin on the lens surface, were seen in the eyes exposed to the nucleic acid samples, and they resolved in 24 hours. Mild (mostly 1+) conjunctival congestion, cells, flare, and fibrin were seen in a few eyes exposed to the 2 denatured OVDs and their controls, with the response durations being shorter in the denatured OVD eyes (24 hours) than in the nondenatured OVD eyes (48 hours). Anterior chamber inflammation was generally observed in fewer denatured OVD eyes than in nondenatured OVD eyes, particularly the dispersive OVD eyes. CONCLUSIONS: Intracameral injection of human albumin protein did not cause ocular inflammation. Nucleic acid intracamerally injected into rabbit eyes caused acute inflammation that quickly resolved. Cohesive and dispersive OVD denatured by drying and steam sterilization alone did not cause ocular inflammation.

Rabbit ocular reactivity to bacterial endotoxin contained in aqueous solution and ophthalmic viscosurgical devices.

OBJECTIVE: To describe the ocular reactivity of the rabbit to bacterial endotoxin contained in an aqueous medium and in a cohesive and a dispersive ophthalmic viscosurgical device (OVD). DESIGN: Experimental, randomized animal study. PARTICIPANTS: Seventy-five New Zealand white rabbits. METHODS: This study was performed using 75 rabbits to evaluate the ocular reactivity to bacterial endotoxin contained in Dulbecco's phosphate-buffered saline (DPBS), a cohesive OVD, and a dispersive OVD. For each test material, 25 rabbits were randomized into 5 groups and were exposed to the test material containing 0.75 endotoxin units (EU), 0.25 EU, 0.08 EU, and 0.02 EU of endotoxin or the vehicle control. The rabbits in each group received bilateral intracameral injection of 0.05 ml of the same test material. All eyes were examined by slit-lamp biomicroscopy at baseline, 3, 6, 9, 24, 48, and 72 hours after injection. At 24 and 72 hours, slit-lamp biomicroscopy (and additionally indirect ophthalmoscopy) was performed through dilated pupils. MAIN OUTCOME MEASURES: Corneal clouding, anterior chamber (AC) flare, cells and fibrin, vitreous haze and cells, cells and fibrin on lens surface, lens opacities, and onset time. RESULTS: The inflammation seen after exposure to the 3 endotoxin-spiked materials followed the same general time course. Anterior chamber cells, flare, iris hyperemia, and conjunctival congestion were seen as early as 3 hours. They started to diminish after 6 hours (DPBS eyes) and 9 hours (OVDs) and were not detectable at 48 and 72 hours, respectively. The AC inflammation was more severe in the OVD eyes than in the DPBS eyes. Anterior chamber fibrin was seen in the OVD eyes only, which persisted through 72 hours in many eyes. A trend toward a dose-response relationship was seen for AC cells and flare and the presence of cells and fibrin on the lens surface in all 3 treatment groups in the first 24 hours. CONCLUSIONS: Inflammation was seen after intracameral injection of as little as 0.02 and 0.08 EU in OVD and DPBS eyes, respectively. Observed responses to intracamerally injected endotoxin in OVDs were more severe and of longer duration than those in aqueous medium.

Detecting endotoxin contamination of ophthalmic viscosurgical devices: intracameral versus intravitreal assays in rabbits.

OBJECTIVE: To compare the sensitivities of intracameral and intravitreal assays in the rabbit model to determine the relative adequacy of these methods in detecting bacterial endotoxin contamination of ophthalmic viscosurgical devices (OVDs). DESIGN: Experimental, randomized animal study. PARTICIPANTS: Twenty New Zealand white rabbits. METHODS: Rabbits were randomized into 4 groups to receive a cohesive or a dispersive OVD via intracameral or intravitreal injection. All 40 treated eyes (10 eyes of 5 animals in each group) received bilateral injection of OVD spiked with bacterial endotoxin at 7.0 endotoxin units/ml. All eyes were evaluated by slit-lamp biomicroscopy for inflammatory response at 3, 6, 9, 24, 48, and 72 hours after exposure. Eyes that received intravitreal injection were also dilated at 24, 48, and 72 hours and were re-examined by slit-lamp biomicroscopy and by indirect ophthalmoscopy. MAIN OUTCOME MEASURES: Conjunctival inflammation, anterior chamber (AC) flare, cells and fibrin, vitreous haze and cells, iridal hyperemia, corneal clouding, lens opacities, and onset times. RESULTS: Intracamerally injected eyes frequently showed conjunctival congestion, AC cells and flare, iridal hyperemia, and fibrin within 6 hours. Up to 80% showed AC cells and flare at 9 hours, and up to 70% showed fibrin at 24 hours. These signs diminished within 48 hours. Fibrin and cells also were seen on the lens surface of most of the eyes. Intravitreally injected eyes showed no signs of inflammation within 24 hours, other than some conjunctival inflammation. After the 24-hour time point, in addition to some conjunctival inflammation, some other signs of inflammation were observed infrequently in the intravitreally injected eyes, including minor vitreous cell reaction in 2 eyes. Although there was 1 dispersive OVD-treated eye with cells and fibrin on the lens capsule at 48 hours, no aqueous cells or flare were seen in the AC of any intravitreally injected eyes at any time during the course of the study. CONCLUSIONS: The rabbit intravitreal assay, when limited to 72 hours, does not seem to have adequate sensitivity to detect endotoxin reliably in OVDs.

Sporicidal Action of Hydrogen Peroxide on Conidia From Toxigenic Strains of Aspergillus flavus and Aspergillus parasiticus.

Effects of pH, sucrose, glucose, and sodium chloride on resistance of 14-day-old conidiospores of Aspergillus parasiticus NRRL 2999 and 3315, and Aspergillus flavus NRRL 3353 to a solution of 6% hydrogen peroxide at 20 C were determined. An increase in time necessary to attain 99.9% destruction of spores resulted when the pH of hydrogen peroxide was adjusted from an initial value of 3.79 to 6.40 and 8.30. However, the amount of the increase was strain-dependent and was directly related to resistance of spores to peroxide. Addition of 10 to 40% sucrose or 3 to 10% sodium chloride to the menstruum caused either an increase or decrease in the time needed to destroy 99.9% of the spore population, depending on the amount of additive that was used. Spores were more difficult to inactivate by peroxide in the presence rather than absence of sucrose or salt, but protection afforded by the additives diminished when their concentration was great. Addition of 10 to 30% glucose elicited a similar response only from spores produced by the most resistant of the three strains tested, whereas resistance of spores from the other strains was not markedly affected.

Sporicidal Action of Hydrogen Peroxide on Conidia From Toxigenic Strains of Aspergillus flavus and Aspergillus parasiticus.

Effectiveness of hydrogen peroxide was evaluated as a sporicidal agent aginst conidia of Aspergillus parasiticus NRRL 2999 and 3315, and Aspergillus flavus NRRL 3353. Conidia were harvested from 7-, 10-, and 14-day-old mold cultures grown on modified Moyer's agar, treated with hydrogen peroxide, and then were recovered with mycological agar. An initial spore concentration of 5 × 105 to 1 × 106 per ml was treated with 2, 4, and 6% (wt/vol) peroxide at 40, 30, and 20 C. Survival curves were not straight logarithmic but tended to tail off at the end. Time required for 99.9% reduction of spores ranged from minutes to an hour, and on rare occasions it took hours to achieve the first logarithmic reduction. Conidia of NRRL 3353 consistently were most resistant among the three strains tested. Conidia of NRRL 2999 and 3315 were equally sensitive to peroxide in most instances, although time required for 99.9% destruction of the latter was occasionally greater. Conidia from 14-, and 10-day-old cultures of NRRL 3315 and 3353, respectively, were more resistant than conidia from the other mold cultures. Resistance of conidia from NRRL 2999 was not affected by age of the culture.