September 2012
Volume 53, Issue 10
Free
Cornea  |   September 2012
Protective Effects of Dispersive Viscoelastics on Corneal Endothelial Damage in a Toxic Anterior Segment Syndrome Animal Model
Author Notes
  • From the Department of Ophthalmology, Korea University College of Medicine, Seoul, South Korea. 
  • Corresponding author: Jong-Suk Song, Department of Ophthalmology, Guro Hospital, Korea University College of Medicine 80, Guro-dong, Guro-gu, Seoul, 152-703, South Korea; crisim@korea.ac.kr
Investigative Ophthalmology & Visual Science September 2012, Vol.53, 6164-6170. doi:10.1167/iovs.12-9945
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jong-Suk Song, Jeong-Hwa Heo, Hyo-Myung Kim; Protective Effects of Dispersive Viscoelastics on Corneal Endothelial Damage in a Toxic Anterior Segment Syndrome Animal Model. Invest. Ophthalmol. Vis. Sci. 2012;53(10):6164-6170. doi: 10.1167/iovs.12-9945.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: We evaluated whether viscoelastics have protective effects on the corneal endothelial cell damage in a toxic anterior segment syndrome (TASS) animal model depending on the types of viscoelastics.

Methods.: A TASS animal model was established with an injection of 0.1 mL o-phthaldehyde solution (0.14%) into the anterior chamber of New Zealand white rabbits. One of two different viscoelastics, 1% sodium hyaluronate (cohesive group) or a 1:3 mixture of 4% chondroitin sulfate and 3% sodium hyaluronate (dispersive group), was injected into the anterior chamber. After five minutes, it was removed using a manual I/A instrument, and then 0.1 mL of o-phthaldehyde solution (0.14%) was injected into the anterior chamber. Damage to corneal endothelial cells was compared between the two groups.

Results.: The corneal thickness increased quickly in both groups after the disinfectant injection. However, the dispersive group showed relatively mild corneal edema compared to the cohesive group. The mean corneal haze score in the dispersive group also was lower than that of the cohesive group. These partial protective effects of the dispersive viscoelastic were demonstrated by the different findings of a live/dead cell assay, TUNEL staining, and scanning electron microscopy between the two groups.

Conclusions.: The TASS animal model seems to be a useful means to evaluate corneal endothelial cell damage caused by toxic substances to find ways to protect or reduce endothelial cell damage. Dispersive viscoelastics were shown to have partial protective effects against corneal endothelial cell damage caused by a toxic disinfectant.

Introduction
Toxic anterior segment syndrome (TASS) is a sterile postoperative inflammatory reaction caused by noninfectious toxic substances that enter the anterior segment during surgery, resulting in toxic damage to intraocular tissues. 1 Many TASS cases have been reported from many countries. Some cases were an outbreak, while other cases were sporadic. 25 The substances that are proven to cause TASS are many, including denatured viscoelastics, disinfectants, preservatives, ointment, plasma gas sterilization, and so forth. 39 Unfortunately, TASS still is a significant issue in cataract surgery because it can induce severe damage to the corneal endothelium and trabecular meshwork, leading to bullous keratopathy and glaucoma. 1  
Previous reports on TASS only dealt with the causes of TASS, clinical findings, and response to treatment because TASS research is very limited in the clinical setting. Therefore, TASS animal models are needed to investigate the effects of toxic substances on various intraocular tissues, such as the corneal endothelium, iris, and trabecular meshwork. 
Many different substances have been related to TASS, and clinical findings and their outcomes quite differ according to the toxic substances. In our study, a TASS animal model was established using a commercially available disinfectant, 0.55% o-phthaldehyde (Cidex OPA; Advanced Sterilization Products, Irvine, CA), which is used popularly for cold sterilization of medical equipment and has been reported to induce TASS after cataract surgery. 8 In addition, we sought to elucidate whether viscoelastics have protective effects on the corneal endothelial cell damage depending on the types of viscoelastics using this TASS animal model. 
Methods
A total of 24 New Zealand white rabbits weighing 2.0 to 3.0 kg each was used in this study. The right eye of each rabbit was selected for the procedure and the left eye was used as the normal control, which was not subject to any surgery or any intraocular injection. The use of the rabbits conformed to the Association for Research in Vision and Ophthalmology's (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. 
A solution of 0.55% o-phthaldehyde was used to establish an animal model with TASS. It was diluted with balanced salt solution (BSS) to a ratio of 1:3 (final concentration: 0.14%), and then 0.1 mL was injected into the anterior chamber in 8 rabbits. The central corneal thickness was measured using an ultrasound corneal pachymeter (BV International, Clermont-Ferrand, France) before and 12 hours after injection. Corneal thickness was measured three times, and the mean values were used for statistical analysis. 
Hematoxylin and eosin staining, a live/dead cell assay, and TUNEL staining were performed to evaluate corneal endothelial damage. Rabbits were anesthetized with an intramuscular injection of ketamine hydrochloride, and were killed humanely with an intracardiac injection of 2% lidocaine hydrochloride. The eyes were enucleated and frozen immediately in an optimum cutting temperature (OCT) compound (Tissue-Tek; Miles Laboratories, Elkhart, IN) with liquid nitrogen. Frozen tissue blocks were stored at −85°C until they were sectioned. Central corneal sections (7 μm thick) were cut using a cryostat at −20°C and placed on silanized microscope slides (DAKO, Carpinteria, CA). Some tissue sections were stained with hematoxylin and eosin for histologic observation, while the others were fixed in cold acetone at −20°C for 2 minutes for TUNEL staining. TUNEL assays were performed in 5 rabbits using the ApopTag Red In Situ Apoptosis Detection Kit (Cat No. S7165; Chemicon International, Temecula, CA). DAPI was used for counterstaining of nuclei. 
To assess the endothelial cell damage, cell viability was evaluated in 3 rabbits using a live/dead viability/cytotoxicity kit (Molecular Probes, Eugene, OR). Staining was performed according to the manufacturer's instructions. Live cells, which are distinguished by the presence of ubiquitous intracellular esterase activity, appear green, whereas dead cells with damaged membranes are stained red. 
To evaluate the effects of viscoelastics on corneal endothelial cell damage in the TASS animal model, two different viscoelastics were compared. One was 1% sodium hyaluronate (Hyalu; Hanmi Pharma Co., Seoul, South Korea), which is a cohesive viscoelastic (cohesive group). The other was a dispersive viscoelastic (dispersive group) containing 4% chondroitin sulfate and 3% sodium hyaluronate (Viscoat; Alcon Laboratories, Inc., Fort Worth, TX). A 2.2 mm corneal incision was made in the right eyes of 16 rabbits, and then the anterior chamber was filled with one of the two viscoelastics. After 5 minutes, the viscoelastic was removed using a manual irrigation and aspiration (I&A) instrument for 120 seconds. Then 0.1 mL of Cidex OPA diluted with BSS to a ratio of 1:3 was injected into the anterior chamber. 
To compare the damage to the corneal endothelium, corneal thickness was measured in 10 rabbits (n = 5 each group) before, and 6 and 24 hours after injection. Corneal clarities were compared between the two groups (n = 5 each group) 24 hours after injection. Corneal clarities were evaluated according to the Fantes' classification: Grade 0, a totally clear cornea; Grade 0.5, trace haze faintly detectable with broad illumination; Grade 1, minimal haze seen easily with broad illumination; Grade 2, mild haze easily visible with direct focal slit illumination; Grade 3, moderate opacity obscuring iris details; and Grade 4, severe opacity blocking the view of anterior chamber structures. 10  
In addition to hematoxylin and eosin staining, the live/dead cell assay, and TUNEL staining, inflammatory cell infiltration was evaluated with a CD11b immunohistochemical assay (n = 3 each group). The appearance of corneal endothelial cells was evaluated using scanning electron microscopy (n = 2 each group). The monoclonal antibody for CD11b (BD Corp, San Diego, CA), a marker of monocytes, was placed on the sections, and incubated at 4°C overnight (anti-CD11b). The working concentration for the CD11b antibody was 1:50 (5% BSA). The secondary antibody for CD11b staining was Alexa Fluor 555 goat anti-rat IgG (Invitrogen Corp., Carlsbad, CA) diluted 1:500, which was applied for 1 hour at room temperature. Coverslips were mounted with Vectashield containing 4′,6 diamidino-2-phenylindole (DAPI; Sigma, St. Louis, MO) to observe all cell nuclei in the tissue. CD11b-positive cells were counted in 4 randomly selected high-power fields per slide (original magnification ×16) and compared between the two groups. In the live/dead cell assay, dead cells also were counted and compared in the same way. For scanning electron microscopy, corneas were prefixed in 2% glutaraldehyde in 0.1 M phosphate buffer and post-fixed for 2 hours in 1% osmic acid dissolved in PBS. Then, they were treated in a graded series of ethanol and t-butyl alcohol, dried in a freeze dryer (ES-2030; Hitachi, Tokyo, Japan), coated with platinum using an ion coater (IB-5; Eiko, Ibaraki, Japan), and observed via FE-SEM (S-4700; Hitachi). 
The Statistical Package for the Social Sciences (SPSS) software (version 10.1; SPSS, Chicago, IL) was used for statistical analyses. A nonparametric Mann-Whitney test was used to compare the mean corneal thickness and corneal haze, and the mean number of dead cells and CD11b-positive cells between the two groups. P values <0.05 were considered statistically significant. 
Results
Injection of 0.1 mL Cidex OPA diluted with BSS to a ratio of 1:3 into the right eyes of 8 rabbits induced severe corneal edema 12 hours after injection. The mean corneal thickness was increased remarkably from 395.9 ± 20.4 μm at baseline to 1150.3 ± 185.4 μm 12 hours after injection. The mean corneal thickness of the normal control was 393.0 ± 18.6 μm before the injection in the opposite eye and 410.2 ± 21.3 μm 12 hours after the injection. The difference between the corneal thicknesses of experimental eyes and control eyes was statistically significant after injection (P < 0.001, Fig. 1). 
Figure 1. 
 
The change of mean corneal thickness before and 12 hours after disinfectant injection
Figure 1. 
 
The change of mean corneal thickness before and 12 hours after disinfectant injection
Hematoxylin and eosin staining showed a very edematous cornea compared to control eyes, and most endothelial cells were lost (Figs. 2A, 2B). In the live/dead cell assay (n = 3 each group) the corneas of the normal eyes showed typical endothelial cell structure and most endothelial cells were live (Fig. 3A). At 12 hours after the disinfectant injection, the normal structure was destroyed severely, and many dead cells were observed (Fig. 3B). TUNEL-positive cells were not detected in the endothelium of normal rabbit corneas (Fig. 4A). However, after the disinfectant injection, many endothelial cells were TUNEL-positive (Fig. 4B, n = 5 each group). Disinfectant injection into the anterior chamber made the rabbit corneas very similar to the corneas of patients with TASS. 7 Therefore, this method seems to be successful for generating a TASS animal model. 
Figure 2. 
 
Hematoxylin and eosin stain. (A) Remarkable stromal edema was observed after disinfectant injection (×40). (B) In contrast with normal cornea (inset), most endothelial cells were lost (×200).
Figure 2. 
 
Hematoxylin and eosin stain. (A) Remarkable stromal edema was observed after disinfectant injection (×40). (B) In contrast with normal cornea (inset), most endothelial cells were lost (×200).
Figure 3. 
 
Live/dead cell assay. (A) Almost all endothelial cells of normal cornea are viable and stained green. (B) At 12 hours after disinfectant injection, many cells turned to red dead cells, and normal endothelial cell structure was destroyed severely.
Figure 3. 
 
Live/dead cell assay. (A) Almost all endothelial cells of normal cornea are viable and stained green. (B) At 12 hours after disinfectant injection, many cells turned to red dead cells, and normal endothelial cell structure was destroyed severely.
Figure 4. 
 
TUNEL stain. (A) In normal rabbit corneas, TUNEL-positive cells are not detected in corneal endothelium. (B) At 12 hours after disinfectant injection, some TUNEL-positive cells (arrows) are observed in the corneal endothelium.
Figure 4. 
 
TUNEL stain. (A) In normal rabbit corneas, TUNEL-positive cells are not detected in corneal endothelium. (B) At 12 hours after disinfectant injection, some TUNEL-positive cells (arrows) are observed in the corneal endothelium.
In the second experiment to evaluate the effect of viscoelastics on corneal endothelial damage in this TASS animal model, rabbit corneas were clear before the disinfectant injection and there was no significant difference between the two groups. However, the corneal haze in the rabbit eyes 24 hours after the disinfectant injection was less severe in the dispersive group than in the cohesive group though (Figs. 5A, 5B). The mean haze score was 2.6 ± 0.7 in the cohesive group, while it was 1.4 ± 0.5 in the dispersive group (P = 0.03). The corneal thickness increased quickly with time in both groups after the disinfectant injection. However, the dispersive group showed relatively milder corneal edema than the cohesive group (n = 5 each group, Table 1). 
Figure 5. 
 
Corneal hazes of rabbit eyes in the two groups at 24 hours after disinfectant injection. (A) Grade 3 corneal haze of rabbit eye in the cohesive group. (B) Grade 1 corneal haze of rabbit eye in the dispersive group.
Figure 5. 
 
Corneal hazes of rabbit eyes in the two groups at 24 hours after disinfectant injection. (A) Grade 3 corneal haze of rabbit eye in the cohesive group. (B) Grade 1 corneal haze of rabbit eye in the dispersive group.
Table 1. 
 
The Change of Mean Central Corneal Thickness after Disinfectant Injection in Both Groups (n = 5 Each Group)
Table 1. 
 
The Change of Mean Central Corneal Thickness after Disinfectant Injection in Both Groups (n = 5 Each Group)
Central Corneal Thickness, mm Pre-Injection 6 Hours after Injection 24 Hours after Injection
Cohesive group 352.3 (± 31.7) 740.8 (± 152.6) 1181.3 (± 265.7)
Dispersive group 348.6 (± 12.7) 558.1 (± 97.4) 790.0 (± 73.3)
P value 0.818 0.041 0.009
In the live/dead cell assay performed 24 hours after the injection (n = 3 each group), many endothelial cells were dead in both groups (Figs. 6A, 6B). However, the mean number of dead cells in the dispersive group (74.0 ± 11.9) was less than that of the cohesive group (84.7 ± 3.2, P < 0.01), and the corneal endothelial structure was partially maintained in the dispersive group (Fig. 6B). TUNEL staining in the cohesive group (n = 3) did not reveal any endothelial cells on the bright field image. TUNEL-positive cells were located beneath Descemet's membrane, which means that they were not corneal endothelial cells, but deep stromal keratocytes (Fig. 7A). However, in the dispersive group (n = 3), corneal endothelial cells were observed on the bright field image, and some of them were TUNEL-positive (Fig. 7B). 
Figure 6. 
 
Live/dead cell assay at 24 hours after injection. Most endothelial cells were dead in the cohesive group (A) as well as in the dispersive group (B). However, corneal endothelial structure was observed partially in the dispersive group (B).
Figure 6. 
 
Live/dead cell assay at 24 hours after injection. Most endothelial cells were dead in the cohesive group (A) as well as in the dispersive group (B). However, corneal endothelial structure was observed partially in the dispersive group (B).
Figure 7. 
 
TUNEL stain at 24 hours after injection. (A) In the cohesive group, endothelial cells are not observed on the bright field image, and TUNEL-positive cells were located beneath Descemet's membrane, which means that they are not corneal endothelial cells, but deep stromal keratocytes (arrows). (B) In the dispersive group, corneal endothelial cells are observed on the bright field image, and some of them are TUNEL-positive (arrows).
Figure 7. 
 
TUNEL stain at 24 hours after injection. (A) In the cohesive group, endothelial cells are not observed on the bright field image, and TUNEL-positive cells were located beneath Descemet's membrane, which means that they are not corneal endothelial cells, but deep stromal keratocytes (arrows). (B) In the dispersive group, corneal endothelial cells are observed on the bright field image, and some of them are TUNEL-positive (arrows).
In CD11b immunohistochemical analysis (n = 3 each group), CD11b–positive cells were not observed in either the center or periphery of normal corneas. However, 24 hours after the disinfectant injection, numerous CD11b-positive cells were observed in the peripheral cornea (Figs. 8A bottom, 8B bottom), whereas in the corneal center, CD11b-positive cells were much less common (Figs. 8A top, 8B top). Inflammatory cell infiltration in the dispersive group was less severe in the corneal center as well as in the peripheral cornea than that of the cohesive group (P < 0.001 and P < 0.001, respectively, Table 2). 
Figure 8. 
 
CD11b immunohistochemistry. (A) In the cohesive group, numerous CD11b-positive cells (arrows) are observed in the peripheral cornea (bottom), whereas CD11b-positive cells are much less in the corneal center (top). (B) In the dispersive group, the pattern of inflammatory cell infiltration is similar to that of the cohesive group. However, inflammatory cell infiltrations (arrows) are less severe in the corneal center (top) as well as in the peripheral cornea (bottom).
Figure 8. 
 
CD11b immunohistochemistry. (A) In the cohesive group, numerous CD11b-positive cells (arrows) are observed in the peripheral cornea (bottom), whereas CD11b-positive cells are much less in the corneal center (top). (B) In the dispersive group, the pattern of inflammatory cell infiltration is similar to that of the cohesive group. However, inflammatory cell infiltrations (arrows) are less severe in the corneal center (top) as well as in the peripheral cornea (bottom).
Table 2. 
 
The Mean Number of CD11b-Positive Cells in Both Groups 24 Hours after Disinfectant Injection (n = 3 Each Group)
Table 2. 
 
The Mean Number of CD11b-Positive Cells in Both Groups 24 Hours after Disinfectant Injection (n = 3 Each Group)
CD11b-Positive Cells Center Periphery
Cohesive group 9.5 (± 1.3) 254.0 (± 22.3)
Dispersive group 3.7 (± 1.0) 147.3 (± 7.4)
P value < 0.001 < 0.001
Scanning electron microscopy (n = 2 each group) showed that normal corneal endothelium displayed a uniform hexagonal appearance with regular, interdigitated cell borders and distinct microvilli on the cell surface (Fig. 9A). However, in the cohesive group, many parts of the endothelial cell layer were detached, and the remaining endothelial cells had lost microvilli on the surface and their intercellular junctions were destroyed (Fig. 9B). Compared to the cohesive group, the endothelial cells in the dispersive group were relatively less damaged (Fig. 9C). 
Figure 9. 
 
Photographs of scanning electron microscopy. (A) Normal corneal endothelium shows a uniform hexagonal appearance with regular interdigitated cell borders and distinct microvilli on the cell surface. (B) In the cohesive group, parts of endothelial cell layer are detached (arrows), and remaining endothelial cells lose microvilli on the surface, and intercellular junctions are destroyed. (C) In the dispersive group, the endothelial cells are relatively less damaged, compared to the cohesive group.
Figure 9. 
 
Photographs of scanning electron microscopy. (A) Normal corneal endothelium shows a uniform hexagonal appearance with regular interdigitated cell borders and distinct microvilli on the cell surface. (B) In the cohesive group, parts of endothelial cell layer are detached (arrows), and remaining endothelial cells lose microvilli on the surface, and intercellular junctions are destroyed. (C) In the dispersive group, the endothelial cells are relatively less damaged, compared to the cohesive group.
Discussion
Two serious early postoperative complications after cataract surgery are infectious endophthalmitis and TASS. As the clinical findings of TASS can differ depending on the toxic substances, differential diagnosis between these two complications is very important in clinical settings. One of the typical clinical findings in TASS is very severe corneal edema from limbus to limbus. This diffuse corneal edema often is not noted in infectious endophthalmitis. TASS corneal edema also is different than the central corneal edema induced by prolonged use of ultrasound energy needed with a dense nuclear cataract. Because toxic substances entering the anterior chamber in TASS do damage to the entire cornea, limbus to limbus corneal edema is seen. 
In our study, Cidex OPA disinfectant solution, which has been proven to cause TASS in a previous case report, induced severe corneal edema 12 hours after an anterior chamber injection. 8 The corneal findings were very similar to those seen in clinical TASS patients. Compared to the corneal thickness in the normal control (410.2 μm), the corneal thickness increased markedly to 1150.3 μm 12 hours after injection. Devastating damage to the corneal endothelium caused by the disinfectant was observed via hematoxylin and eosin staining, a live/dead cell assay, and TUNEL staining. The corneas were very edematous with severely destroyed endothelial cell structure, many dead cells, and many TUNEL-positive endothelial cells. This disinfectant, which was reported clinically to cause TASS, induced severe damage to the rabbit corneal endothelium. Therefore, this TASS animal model seems to be useful for evaluating corneal endothelial damage as well as damage to other intraocular tissues, and for finding effective ways to protect or reduce damage to corneal endothelial cells. 
First, we tried to elucidate whether viscoelastics have any protective effects against corneal endothelial cell damage depending on the type of viscoelastics used in this TASS animal model. Generally, viscoelastics are classified into two types, cohesive and dispersive, based on their characteristics. Cohesive viscoelastics are better in maintaining the anterior chamber and are easier to remove from the eye. However, dispersive viscoelastics are better in protecting the corneal endothelium from damage during cataract surgery. 11 We wondered whether dispersive viscoelastics could protect the corneal endothelium from the disinfectant in our TASS animal model. In this study, as we remove viscoelastics at the end of cataract surgery, we washed viscoelastics out before the disinfectant injection. Although the time needed to wash all viscoelastics out is different based on their characteristics, we could remove almost all viscoelastics using manual I/A instrument within 120 seconds under surgical microscopy. To standardize the viscoelastic removal, we washed the anterior chamber with BSS for 120 seconds in each rabbit eye. In contrast with cohesive viscoelastics, dispersive viscoelastics may adhere to the corneal endothelium and remain in the anterior chamber. The different characteristics of dispersive viscoelastics seem to have a role in protecting the corneal endothelium from the toxic substance. 
In our study, the corneal thickness quickly increased with time in both groups after the disinfectant injection. However, the dispersive group showed milder corneal edema than the cohesive group. The mean corneal haze score in the dispersive group also was lower than that of the cohesive group. These partial protective effects of the dispersive viscoelastics were supported by the different findings of the live/dead cell assay, TUNEL staining, and scanning electron microscopy between the two groups. As a result, inflammatory cell infiltration in the central and peripheral cornea was less severe in the dispersive group than in the cohesive group. We demonstrated clearly that dispersive viscoelastics exhibited partial protective effects in corneal endothelial cells against damage caused by the toxic disinfectant when compared to cohesive viscoelastics. However, even in the dispersive group, corneal edema was remarkable, and corneal endothelial damage was severe. Therefore, further studies are needed to find more effective ways to protect against or reduce corneal endothelial damage. 
A limitation in this TASS model is that rabbits were used. Although rabbit animal models have several advantages, rabbit corneal endothelial cells are different than human corneal endothelial cells with respect to cell proliferation. Rabbit corneal endothelial cells can proliferate in vivo. Van Horn et al. reported that autoradiographic studies at 24 hours after transcorneal freezing injury to rabbit corneas revealed many endothelial cells at the margin of the wound took up tritiated thymidine into nuclear DNA, indicating that cellular division was going to occur in these cells. 12 In rabbits with 10% endothelial cell destruction from freezing injury, corneal thickness returned to normal thickness by day 13, and in the 50% and 90% cell destruction group, corneal thickness returned to a near normal value by day 28. Therefore, this TASS animal model is not appropriate for evaluating the long term effects of a disinfectant on the corneal endothelial cells of a human. However, this TASS animal model seems useful for evaluating toxic damage to corneal endothelial cells over a short period of time. 
In conclusion, a TASS animal model was established, and it appears to be a useful way to evaluate corneal endothelial cell damage caused by toxic substances, and to find ways to protect against or reduce endothelial cell damage. Dispersive viscoelastics showed partial protective effects against corneal endothelial cell damage caused by a toxic disinfectant when compared to cohesive viscoelastics. 
References
Mamalis N Edelhauser HF Dawson DG Chew J LeBoyer RM Werner L. Toxic anterior segment syndrome. J Cataract Refract Surg . 2006;32:324–333. [CrossRef] [PubMed]
Hellinger WC Hasan SA Bacalis LP Outbreak of toxic anterior segment syndrome following cataract surgery associated with impurities of autoclave steam moisture. Infect Control Hosp Epidemiol . 2006;27:294–298. [CrossRef] [PubMed]
Duffy RE Brown SE Caldwell KL An epidemic of corneal destruction caused by plasma gas sterilization. The Toxic Endothelial Cell Destruction Syndrome Investigative Team. Arch Ophthalmol . 2000;118:1167–1176. [CrossRef] [PubMed]
Werner L Sher JH Taylor JR Toxic anterior segment syndrome and possible association with ointment in the anterior chamber following cataract surgery. J Cataract Refract Surg . 2006;32:227–235. [CrossRef] [PubMed]
Liu H Routley I Teichmann KD. Toxic endothelial cell destruction from intraocular benzalkonium chloride. J Cataract Refract Surg . 2001;27:1746–1750. [CrossRef] [PubMed]
Kim JH. Intraocular inflammation of denatured viscoelastic substance in cases of cataract extraction and lens implantation. J Cataract Refract Surg . 1987;13:537–542. [CrossRef] [PubMed]
Unal M Yücel I Akar Y Oner A Altin M. Outbreak of toxic anterior segment syndrome associated with glutaraldehyde after cataract surgery. J Cataract Refract Surg . 2006;32:1696–1701. [CrossRef] [PubMed]
Jun EJ Chung SK. Toxic anterior segment syndrome after cataract surgery. J Cataract Refract Surg . 2010;36:344–346. [CrossRef] [PubMed]
Eleftheriadis H Cheong M Sandeman S Corneal toxicity secondary to inadvertent use of benzalkonium chloride preserved viscoelastic material in cataract surgery. Br J Ophthalmol . 2002;86:299–305. [CrossRef] [PubMed]
Fantes FE Hanna KD Waring GO 3rd Pouliquen Y Thompson KP Savoldelli M. Wound healing after excimer laser keratomileusis (photorefractive keratectomy) in monkeys. Arch Ophthalmol . 1990;108:665–675. [CrossRef] [PubMed]
Koch DD Liu JF Glasser DB Merin LM Haft E. A comparison of corneal endothelial changes after use of Healon or Viscoat during phacoemulsification. Am J Ophthalmol . 1993;115:188–201. [CrossRef] [PubMed]
Van Horn DL Sendele DD Seideman SS Buco PJ. Regenerative capacity of the corneal endothelium in rabbit and cat. Invest Ophthalmol Vis Sci . 1977;16:597–613. [PubMed]
Footnotes
 Supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science, and Technology (2010-0007817). The authors alone are responsible for the content and writing of the paper.
Footnotes
 Disclosure: J.-S. Song, None; J.-H. Heo, None; H.-M. Kim, None.
Figure 1. 
 
The change of mean corneal thickness before and 12 hours after disinfectant injection
Figure 1. 
 
The change of mean corneal thickness before and 12 hours after disinfectant injection
Figure 2. 
 
Hematoxylin and eosin stain. (A) Remarkable stromal edema was observed after disinfectant injection (×40). (B) In contrast with normal cornea (inset), most endothelial cells were lost (×200).
Figure 2. 
 
Hematoxylin and eosin stain. (A) Remarkable stromal edema was observed after disinfectant injection (×40). (B) In contrast with normal cornea (inset), most endothelial cells were lost (×200).
Figure 3. 
 
Live/dead cell assay. (A) Almost all endothelial cells of normal cornea are viable and stained green. (B) At 12 hours after disinfectant injection, many cells turned to red dead cells, and normal endothelial cell structure was destroyed severely.
Figure 3. 
 
Live/dead cell assay. (A) Almost all endothelial cells of normal cornea are viable and stained green. (B) At 12 hours after disinfectant injection, many cells turned to red dead cells, and normal endothelial cell structure was destroyed severely.
Figure 4. 
 
TUNEL stain. (A) In normal rabbit corneas, TUNEL-positive cells are not detected in corneal endothelium. (B) At 12 hours after disinfectant injection, some TUNEL-positive cells (arrows) are observed in the corneal endothelium.
Figure 4. 
 
TUNEL stain. (A) In normal rabbit corneas, TUNEL-positive cells are not detected in corneal endothelium. (B) At 12 hours after disinfectant injection, some TUNEL-positive cells (arrows) are observed in the corneal endothelium.
Figure 5. 
 
Corneal hazes of rabbit eyes in the two groups at 24 hours after disinfectant injection. (A) Grade 3 corneal haze of rabbit eye in the cohesive group. (B) Grade 1 corneal haze of rabbit eye in the dispersive group.
Figure 5. 
 
Corneal hazes of rabbit eyes in the two groups at 24 hours after disinfectant injection. (A) Grade 3 corneal haze of rabbit eye in the cohesive group. (B) Grade 1 corneal haze of rabbit eye in the dispersive group.
Figure 6. 
 
Live/dead cell assay at 24 hours after injection. Most endothelial cells were dead in the cohesive group (A) as well as in the dispersive group (B). However, corneal endothelial structure was observed partially in the dispersive group (B).
Figure 6. 
 
Live/dead cell assay at 24 hours after injection. Most endothelial cells were dead in the cohesive group (A) as well as in the dispersive group (B). However, corneal endothelial structure was observed partially in the dispersive group (B).
Figure 7. 
 
TUNEL stain at 24 hours after injection. (A) In the cohesive group, endothelial cells are not observed on the bright field image, and TUNEL-positive cells were located beneath Descemet's membrane, which means that they are not corneal endothelial cells, but deep stromal keratocytes (arrows). (B) In the dispersive group, corneal endothelial cells are observed on the bright field image, and some of them are TUNEL-positive (arrows).
Figure 7. 
 
TUNEL stain at 24 hours after injection. (A) In the cohesive group, endothelial cells are not observed on the bright field image, and TUNEL-positive cells were located beneath Descemet's membrane, which means that they are not corneal endothelial cells, but deep stromal keratocytes (arrows). (B) In the dispersive group, corneal endothelial cells are observed on the bright field image, and some of them are TUNEL-positive (arrows).
Figure 8. 
 
CD11b immunohistochemistry. (A) In the cohesive group, numerous CD11b-positive cells (arrows) are observed in the peripheral cornea (bottom), whereas CD11b-positive cells are much less in the corneal center (top). (B) In the dispersive group, the pattern of inflammatory cell infiltration is similar to that of the cohesive group. However, inflammatory cell infiltrations (arrows) are less severe in the corneal center (top) as well as in the peripheral cornea (bottom).
Figure 8. 
 
CD11b immunohistochemistry. (A) In the cohesive group, numerous CD11b-positive cells (arrows) are observed in the peripheral cornea (bottom), whereas CD11b-positive cells are much less in the corneal center (top). (B) In the dispersive group, the pattern of inflammatory cell infiltration is similar to that of the cohesive group. However, inflammatory cell infiltrations (arrows) are less severe in the corneal center (top) as well as in the peripheral cornea (bottom).
Figure 9. 
 
Photographs of scanning electron microscopy. (A) Normal corneal endothelium shows a uniform hexagonal appearance with regular interdigitated cell borders and distinct microvilli on the cell surface. (B) In the cohesive group, parts of endothelial cell layer are detached (arrows), and remaining endothelial cells lose microvilli on the surface, and intercellular junctions are destroyed. (C) In the dispersive group, the endothelial cells are relatively less damaged, compared to the cohesive group.
Figure 9. 
 
Photographs of scanning electron microscopy. (A) Normal corneal endothelium shows a uniform hexagonal appearance with regular interdigitated cell borders and distinct microvilli on the cell surface. (B) In the cohesive group, parts of endothelial cell layer are detached (arrows), and remaining endothelial cells lose microvilli on the surface, and intercellular junctions are destroyed. (C) In the dispersive group, the endothelial cells are relatively less damaged, compared to the cohesive group.
Table 1. 
 
The Change of Mean Central Corneal Thickness after Disinfectant Injection in Both Groups (n = 5 Each Group)
Table 1. 
 
The Change of Mean Central Corneal Thickness after Disinfectant Injection in Both Groups (n = 5 Each Group)
Central Corneal Thickness, mm Pre-Injection 6 Hours after Injection 24 Hours after Injection
Cohesive group 352.3 (± 31.7) 740.8 (± 152.6) 1181.3 (± 265.7)
Dispersive group 348.6 (± 12.7) 558.1 (± 97.4) 790.0 (± 73.3)
P value 0.818 0.041 0.009
Table 2. 
 
The Mean Number of CD11b-Positive Cells in Both Groups 24 Hours after Disinfectant Injection (n = 3 Each Group)
Table 2. 
 
The Mean Number of CD11b-Positive Cells in Both Groups 24 Hours after Disinfectant Injection (n = 3 Each Group)
CD11b-Positive Cells Center Periphery
Cohesive group 9.5 (± 1.3) 254.0 (± 22.3)
Dispersive group 3.7 (± 1.0) 147.3 (± 7.4)
P value < 0.001 < 0.001
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×