Endophthalmitis is an inflammatory reaction of intraocular tissue due to infection with pathogenic microorganisms, which can be a complication after cataract surgery and can cause permanent blindness. ¹ The most common microorganisms that cause endophthalmitis after cataract surgery are coagulase-negative staphylococci (33–77%) and Staphylococcus aureus (10–21%), and very rarely gram-negative organisms such as Pseudomonas sp. if an outbreak occurs. ²

To reduce the risk of bacterial contamination in the aqueous humor and the subsequent development of endophthalmitis, the natural flora of the eye, eyelids, and eyelashes can be decreased prior to surgery through the use of intracameral antibiotics. French data from 2008 found that aspiration of the anterior chamber fluid during surgery revealed that bacterial contamination occurs in 2–46% of cataract surgeries using the phacoemulsification technique. ^{3 , 4}

Intracameral antibiotic injection has become a favorable method worldwide even though it is still controversial due to a lack of level 1 evidence. ⁵ There is no standard prophylactic method of post-ocular surgery endophthalmitis. Intracameral antibiotic injection aims to achieve a high concentration of intraocular antibiotics in a short time to eliminate pathogenic microorganisms. ⁶

Endophthalmitis after cataract surgery is less common in countries where intracameral antibiotic injection is the standard method of prophylaxis. Before the use of intracameral cefuroxime, the incidence of endophthalmitis after cataract surgery was between 0.3 to 1.2%, and after its use incidents decreased to only 0.014 to 0.08 (January 2002 and December 2004, in Sweden) %. ^{7 , 8}

The intracameral injection has several advantages; it easy to do, reduces the use of topical antibiotics, and addresses compliance issues with post-operative drug use by patients. ^{9 , 10} Fluoroquinolone has great potential as a prophylactic intracameral antibiotic injection as it has a wide spectrum, good penetration into the ocular tissue, and only mild side effects. ^{6 , 8 , 11}

Levofloxacin is available commercially in Indonesia in the preparation of 5ml bottle of eye drops (Cravit ^®, Santen Pharmaceutical, Japan). Levofloxacin 0.5% eye drops in 5 ml bottle has been clinically proven safe for intracameral injection in rabbits and human eyes undergoing cataract surgery. ^{6 , 12} Off-label intracameral levofloxacin 0.5% injection (Cravit ^®) has been chosen as a standard prophylactic method in the Ophtalmology Department of Cipto Mangunkusumo Hospital. Levofloxacin 0.5% is also commercially available in a single dose 0.6 ml of sterile eye drops preservatives free (LFX ^®, Cendo Pharmaceutical, Bandung, Indonesia), which is a local product in Indonesia. Currently, a commercially available single dose 0.6 ml of levofloxacin 0.5% eye drops has not been accepted as an intracameral antibiotic for humans because there was no safety study yet.

The objective of this study was to evaluate the safety of intracameral injection of levofloxacin 0.5% eye drop single dose 0.6 mL preservative-free (LFX ^®) on rabbit eye. Rabbits were used as experimental animals in this study because certain parts of their eyes bear some similarities to those of humans and are easily observable. The rabbit’s eye globe and cornea are relatively large in size. Upon histopathological examination, the corneal epithelial layer comprises six layers of stratified squamous epithelium, with the deepest layer formed by cylindrical cells, similar to that of the human cornea. Similarly, the rabbit’s vitreous humor has a composition similar to that of humans, consisting of 99% water and 1% hyaluronic acid, resulting in a gel-like consistency.

This study implemented an experimental design, where the animals were randomly allocated into three groups. The research was carried out at the Biomedical Research Center in the Health Research and Development Agency of the National Center General - Cipto Mangunkusumo Hospital animal laboratory, between April and June 2016.

Off-label intracameral injection of antibiotics is very important to know the risk of toxicity. One way that can be done to find out the risk of toxicity is through non-clinical tests on experimental animals. ¹⁴ This study assessed the toxic effect from the structural and functional aspects of the levofloxacin intracameral injection. Functional aspects were assessed by the clinical score and structural aspects were assessed by the histopathology score. Comparison of clinical scores on the 1 ^st, 3 ^rd, 5 ^th and 7 ^th day post-injection showed no statistically significant differences between the three treatment groups. A study by Choi JA, et al. (2009) ⁶ also showed the result of clinical scores in the group of levofloxacin 0.5% did not differ statistically significant compared to the control group (balanced salt solution).

Maximum clinical changes post-injection were recorded in the form of mild corneal opacities or the presence of cells and flare light in the anterior chamber. Mild cells and flare were found in the eyes in the control group but not found in the two treatment groups. Kowalski RP, et al. (2015) ¹⁵ also reported mild clinical changes in the eyes of animal rabbits on the first day after intracameral injection of balanced salt solution. Clinical changes can also occur as a result of treatment and injured corneal. Corneal opacities can be caused by corneal edema. Corneal edema may occur as a result of the inflammatory process and/or a decrease in endothelial function. Inflammation contained in the cornea can be seen as inflammatory cell infiltrates (white lesions on the cornea). The decrease in endothelial function can be caused by the toxic effects of drugs or invasive treatments that damage the endothelial cells lining the cornea.

Invasive treatment in this study had been minimized by the use of 30G needles and injection locations near the superior limbus so as not to damage the central cornea. The toxic effects of antibiotics on corneal endothelial cells were known to occur until 2 hours post-exposure and the endothelial cell death rate was still high, although the drug concentration had been decreased. ⁶ Changes in corneal thickness can be measured quantitatively, but in this study it was not done.

A study by Kim SY, et al. (2018) ¹⁶ showed that central corneal thickness changes in the eyes of rabbits injected with 0.1 ml of levofloxacin 0.5% intracameral were not significantly different compared with the control group (BSS) since the first day post-injection. In the group of levofloxacin 0.5%, the average increase in corneal thickness was 9.00 ± 6.87 μm, whereas in the BSS group amounted to 14.83 ± 28.04 μm. Half of the volume aquos humor was eliminated in 46 minutes and finished in 2 hours. The concentration of levofloxacin 0.5% injected was 200 mg/ml in aquos humor after 2 hours. ¹⁷ The residual concentration, in theory, might be due to their mechanism of bonding with proteins in aqueous humor, and re-release process by surrounding tissues (cornea, iris and ciliary body). ¹⁸

Since the first post-injection day, all treatment groups’ vitreous and retinal appearances were normal. Abnormalities such as vascular changes, vitreous hemorrhage, retinal cotton wool spots, and pale area on the retina were not found. Uda et al. found that levofloxacin at a dose of 500 mg/ml was injected intravitreal directly did not cause signs of toxicity and no change of electroretinogram waves on rabbit retina was observed for 4 weeks. ¹⁷

Factors other than the concentrations that could affect the toxic effects of drugs are pH and osmolality. Corneal endothelial cells can be damaged if they are exposed to conditions that are outside of the pH range of 6.5 to 8.5 and the osmolality range of 200–400 mOsm/L. ^{19 , 20} BSS solution was formulated approaching physiological conditions with pH 7.5 and osmolality between 322–324 mOsm/L so it is safe to use intracameral without damaging the function of endothelial cells and does not induce an inflammatory response. Based on the products information, both preparations of levofloxacin 0.5% eye drops used in this study are both known to have a pH of 6.2 to 6.8 and osmolality 300 mOsm/L, but not with other deposits that are the secret formulation of the drug factory. ¹⁵ The incidence of a severe toxic anterior segment syndrome (TASS) has been reported after the use of intracameral moxifloxacin 0.5% eye drop. ¹⁰ Detergents and mucolytics are known to be the causes of severe TASS. This suggests there are other factors to consider in assessing the safety of intracameral injection ophthalmic antibiotic preparations.

The assessment methods of clinical scores in this study were semi-quantitative and relied on the ability of the examiner to observe clinical changes. Corneal opacities in clinical scores have not yet distinguished the detail and clinical changes due to toxic or inflammatory effects. ⁶ Another examination method is using specular microscopy to measure the corneal thickness, and to assess the density, variability, and hexagonality of corneal endothelial, which provides objective data of the structure and function of the cornea in-vivo. Unfortunately we did not have a specular microscope to carry out this examination.

Histopathology examination showed the presence of corneal endothelial cell cytoplasm vacuolization in more than 50% of the endothelial layer in all groups. Vacuoles appeared uniformly without any endothelial cell found missing. Vacuoles formed inside the cytoplasm are large enough to push the cell nucleus to the edge of the cell wall. Vacuolization of endothelial cells in this study could be said as not significant as a result of the toxic effects of drugs, and it also occurred in the control group. Vacuolization of corneal endothelial cells can be found after an 11 minute post-mortem on the cornea that is not fixed. The vacuolization continues to grow over time until the cell wall lysis occurs within 72 hours post-mortem. ²¹

Previous studies have used different methods to assess the toxic effects of drugs on endothelial cells to avoid factors such as post mortem changes. Choi et al. (2009) ⁶ used an electron microscope to find microstructural changes in the hexagonal shape of corneal endothelial cells exposed to the levofloxacin 0.5% after 7 days of intracameral injection.

Results of the histopathological assessment of ciliary body, retina, and choroid showed no sign of damage or inflammation caused by toxic effects in this study. Concentrations of levofloxacin ophthalmic preparations are still secure at a maximum of 3% (3000 mg/ml). Levofloxacin 10% (10,000 μg/ml) can damage tissue structures due to the toxic effect on the ciliary body in the form of edema and inflammation. ^{17 , 22} In, ¹⁷ measured penetration levels of 0.5% levofloxacin ophthalmic preparations were injected intracameral. They found that after 2 hours, levofloxacin concentrations reached 30 mg/g in the anterior vitreous, 15 ug/g in the equator vitreous, and 10 μg/g in the posterior vitreous. Uda et al. (2014) ¹⁷ found that a dose of levofloxacin 500 mg/ml injected intravitreally did not cause retinal morphological changes and damage in experimental rabbits.

The limitations of this study are the subjectivity of clinical assessments, the limited aspects assessed, and the histopathology score being a modification score. To confirm the diagnosis of retinal disorders, it is essential to perform functional assessments such as visual evoked potentials and electroretinograms, as structural abnormalities of the retina do not always correlate with decreased visual function. Using fixation with a 10% formalin buffer solution does not prevent post-mortem changes in endothelial cells that rapidly occur as endothelial vacuolization. These differences need to be completed and confirmed by examination of endothelial cell function by specular microscopy examination.

In conclusion, the safety of 0.1 ml intracameral injection of levofloxacin 0.5% eye drops single dose 0.6 ml preservative-free is similar to 0.1 ml intracameral injection of levofloxacin 0.5% eye drops 5 ml bottle preservative-free in rabbit eye observed from the clinical score of the cornea, anterior chamber, and vitreous as well as the histopathological changes of the cornea, anterior chamber, ciliary body, vitreous, retina, and choroid.