Triamcinolone對眼部組織細胞毒性之實驗性研究

行政院國家科學委員會專題研究計畫 成果報告

Triamcinolone 對眼部組織細胞毒性之實驗性研究

計畫類別： 個別型計畫

計畫編號： NSC94-2314-B-006-100-

執行期間： 94 年 08 月 01 日至 95 年 07 月 31 日 執行單位： 國立成功大學醫學系眼科

計畫主持人： 張義昇 共同主持人： 曾順輝

計畫參與人員： 吳昭良，曾詩雅，陳美鳳

報告類型： 精簡報告

處理方式： 本計畫可公開查詢

中 華 民 國 95 年 9 月 27 日

Triamcinolone acetonide suspension toxicity to corneal endothelial cells

Yi-Sheng Chang, MD, Shih-Ya Tseng, Sung-Huei Tseng, MD, Chao-Liang Wu, PhD, Mei-Feng Chen, MS

PURPOSE: To investigate the cytotoxicity of triamcinolone acetonide (TA) suspensions to corneal en- dothelial cells (CECs).

SETTING: Department of Ophthalmology, Institute of Clinical Medicine, College of Medicine, National Cheng Kung University, Tainan, Taiwan.

METHODS: New Zealand white rabbit CECs were exposed for 1 minute to balanced salt solution (BSS);

commercial TA suspension (cTA); vehicle-removed TA (–vTA); pure vehicle (V); 1/10 dilutions of cTA, –vTA, or V in BSS; or benzyl alcohol (BA) (cTA preservative) 9 mg/mL. Corneal endothelial cell toxicity was assessed by light microscopy (trypan blue staining) and transmission electron microscopy.

The effects of 3-, 10-, or 30-minute exposures to 1/10 cTA, 1/10 –vTA, or V were also investigated.

RESULTS: One-minute exposures to –vTA or 1/10 –vTA did not damage CECs; however, cTA, V, or 1/10 dilutions of cTA or V caused damage and cells exposed to BA showed severe ultrastructural damage/lysis. A 30-minute exposure to 1/10 –vTA did not cause significant cell damage, whereas 3- to 30-minute exposures to 1/10 cTA or V showed significant time-dependent cytotoxicity.

CONCLUSIONS: Commercial TA suspension was cytotoxic to cultured rabbit CECs because of the pre- servative, BA, in the vehicle. Because 1/10 –vTA appeared to be safe for up to 30 minutes of exposure, use of 1/10 dilutions of vehicle-removed TA is suggested to help surgeons visualize prolapsed vitreous during anterior vitrectomy in complicated cataract surgeries.

J Cataract Refract Surg 2006; 32:1549–1555 Q 2006 ASCRS and ESCRS

Vitreous loss during cataract surgery occurs in 1.4% to 7%

of cases

and may be associated with severe intraoperative and postoperative complications such as a dropped lens fragment, vitreous wick syndrome, cystoid macular edema (CME), and retinal detachment.

To help surgeons visual- ize prolapsed vitreous in the anterior chamber, Burk et al.

introduced a synthetic glucocorticoid, triamcinolone ace- tonide (TA), prepared as a milky-looking suspension in its 10-fold dilution. A TA suspension can also aid in visual- ization of the vitreous and posterior hyaloid during pars plana vitrectomy.

Experimental and clinical studies

report that the ve- hicle or preservative in these suspensions, but probably not the crystalline corticosteroid agent itself, may be toxic to the lens and retina. However, no study has examined the ef- fects of TA suspensions on the corneal endothelium. When preparations are used intracamerally, there is particular concern about their potential to damage the corneal endo- thelium because endothelial damage and subsequent de- compensation can lead to vision-threatening bullous keratopathy.

We previously evaluated the cytotoxicity of dyes and intracameral anesthetic agents on cultured rabbit corneal endothelial cells (CECs) and yielded valid results for cata- ract surgery.

In the study reported here, we used the same experimental model to investigate the potential for a commercial TA suspension to damage CECs and propose a safer preparation for clinical use.

MATERIALS AND METHODS

New Zealand white rabbits were used in this study. All animal procedures were approved by the Institutional Review Board of National Cheng Kung University and were performed in accor- dance with ARVO Statement for Use of Animals in Ophthalmic and Vision Research.

Most biologic products used in this study have been de- scribed.

Those not previously described were commercial TA suspension (cTA, Kenacort-A, 40 mg/1 mL), benzyl alcohol (BA), and Millipore filters (Acrodisc 32 mm syringe filter with 0.2 mm support membrane, Pall Corp.).

Six test solutions were used in this study (Table 1), prepared as follows. Solution 1 was the commercial TA suspension (cTA).

Solution 2 was a vehicle-removed suspension of TA in balanced

salt solution (BSS) (–vTA). Solution 3 was the pure vehicle of cTA (V). Solutions 4, 5, and 6 were 1/10 dilutions of solutions 1, 2, and 3, respectively, in BSS. To prepare solutions 2 and 3, 12 mL of cTA was centrifuged (4000 rpm for 10 minutes) to separate the TA par- ticles from the vehicle.

The sediment of TA particles (about 1 mL) was resuspended in BSS to achieve the initial volume of 12 mL. Centrifugation and resuspension in BSS were repeated once to obtain solution 2 (–vTA). Then, the supernatant of vehicle (approximately 11 mL) from the first centrifugation of cTA was passed through 0.2 mm pore Millipore filters to obtain solution 3 (V), consisting of pure vehicle without TA particles. Diluting 1 mL of solution 1, 2, or 3 with 9 mL of BSS yielded solution 4 (1/10 cTA), solution 5 (1/10 –vTA), and solution 6 (1/10 V), respectively.

To further clarify the effects cTA might have on CECs, a labo- ratory preparation of the main ingredient of the vehicle, BA, was diluted with BSS to achieve a concentration, 9 mg/mL, that was identical to the concentration of BA in cTA.

Cell Culture

New Zealand white rabbits weighing 1.0 to 1.5 kg were killed by intravenous injection of phenobarbital, and CECs were ob- tained from the excised corneas using a method that has been de- scribed.

The cells from 8 animals (16 corneas) were combined

to inoculate 8, 3 cm culture dishes and were incubated in a 5% car- bon dioxide–humidified atmosphere at 37

C. On day 3, the me- dium was changed and Descemet’s membrane fragments were discarded. When cells reached confluence (about day 7), cells in each dish were divided into 4 or 6 portions (depending on the size of the cell monolayer) and subcultured in the same medium in 3 cm dishes. Subcultures (passage 1) grew to confluence in about 4 days and were subcultured again in the same medium in 3 cm culture dishes (passage 2). The cell monolayers that grew from passage 2 were used for experiments.

Exposure to Solutions

In the first experiment, cultures of CECs were exposed to the control (BSS) or test solution by removing the culture medium from the dish and adding 0.5 mL of the solution to overlay the monolayer of endothelial cells. After 1 minute, the solution was carefully washed from the cell culture and the procedure to deter- mine percentage of cells damaged was begun. Each exposure was performed in quadruplicate and the mean used for analysis.

In the second experiment, the cytotoxicity of each solution was evaluated over time by exposing cell cultures to BSS, V, 1/10 cTA, or 1/10 –vTA for 1 (control) for 3, 10, or 30 minutes. Each exposure was performed in quadruplicate and the mean used for analysis.

Determination of Percentage of Damaged Cells

The trypan blue staining method of Chang et al.

was used to determine the percentage of cells in each CEC culture that were damaged by exposure to solutions. After CEC exposure, the con- fluent layer of cells was washed with phosphate-buffered saline (PBS) and trypsinized. Trypsinization was halted by addition of fe- tal bovine serum to the cell suspension to a final concentration of 10%. Trypan blue 0.2% was added for 5 minutes to stain dead and dying cells.

The numbers of stain-positive and stain-negative cells in each culture were counted using a hemocytometer chamber. The mean of the percentages of stain-positive cells in cultures in each exper- imental group was compared to the mean in the control cultures for that experiment. Differences were evaluated using a 2-tailed Student t test. A P value less than 0.05 was considered statistically significant.

Table 1. Test solutions prepared from the commercial triamcinolone acetonide suspension (Kenacort-A) (cTA) or BA.

Group Solution

TA (mg/mL)

Preservative (BA) (mg/mL)

Osmolality Measured* (mOsm/kg)

pH Measured

Control BSS 0 0 305 7.4

Solution 1 Commercial TA (cTA) 40 9 326 6.2

Solution 2 Vehicle-removed TA in BSS (–vTA) w40 0 295 7.7

Solution 3 Pure vehicle (V) 0 9 321 6.1

Solution 4 1/10 dilution of cTA in BSS (1/10 cTA) 4 0.9 306 7.5

Solution 5 1/10 dilution of –vTA in BSS (1/10 –vTA) w4 0 298 7.3

Solution 6 1/10 dilution of V in BSS (1/10 V) 0 0.9 305 6.8

Benzyl alcohol BA 9 mg/mL in BSS 0 9 306 7.4

BA Z benzyl alcohol; BSS Z balanced salt solution

*Osmolality measured by Osmometer, model 3D3 (Advanced Instruments)

†

pH measured by pH meter pH211 (Hanna Instruments)

Accepted for publication April 18, 2006.

From the Department of Ophthalmology (Chang, S.Y. Tseng, S.H.

Tseng), Institute of Clinical Medicine (Chang), Department of Biochemistry and Molecular Biology (Wu), and Institute of Basic Medical Sciences (Chen), College of Medicine, National Cheng Kung University, Tainan, Taiwan.

Supported by grants NCKUH-94-052 from the National Cheng Kung University Hospital, Tainan, and NSC 94-2314-B-006-100 from the National Science Council, Taiwan.

No author has a financial or proprietary interest in any material or method mentioned.

Corresponding author: Sung-Huei Tseng, MD, Department of Ophthalmology, College of Medicine, National Cheng Kung University, 138 Sheng-Li Road, Tainan, Taiwan. E-mail: shtseng1@

gmail.com.

LABORATORY SCIENCE: CORNEAL ENDOTHELIAL CELLS AND TRIAMCINOLONE TOXICITY

J CATARACT REFRACT SURG - VOL 32, SEPTEMBER 2006

1550

Structural Changes in Rabbit Corneal Endothelial Cells When counted in the hemocytometer chamber, cells were also evaluated by light microscopy (LM) for their size and shape and for the integrity of the cell membrane, cytoplasm, and nu- cleus. Ultrastructural changes were evaluated by transmission electron microscopy (TEM) (JEM-1200EX, JEOL). The CECs were cultured in 4-slide chambers and exposed for 1 minute to BSS, cTA, V, or BA. The cell monolayer was washed with PBS and, without trypsinization, was fixed with 2.5% glutaralde- hyde–0.5% paraformaldehyde for 4 hours. After they were rinsed in 0.1 M PBS, cells were osmicated, dehydrated, and embedded in epoxy resin and examined by TEM.

RESULTS

Damage to Corneal Endothelial Cells from 1-Minute Exposures

In the control group, 5.9% of cells stained with trypan blue and stained cells had the same size and shape as un- stained cells (Figure 1, A). Cultures exposed for 1 minute to –vTA or 1/10 –vTA showed relatively intact cellular mor- phology (Figure 1, C and F). In contrast, the trypan blue–

stained cells in cultures exposed for 1 minute to cTA or V showed serious abnormalities (variation in size, irregular cell membranes, cell swelling, and cell lysis) (Figure 1, B, D, E, and G). The CECs exposed to BA 9 mg/mL were so extensively damaged that the percentage of cells damaged could not be determined by viewing in the hemocytometer (Figure 1, H).

The percentages of damaged cells were similar in control cultures and cultures exposed to –vTA (6.9%; P Z .23) or 1/10 –vTA (6.4%; P Z.54) ( Figure 2). In contrast, cultures exposed to solutions with full-strength or diluted vehicle had significantly higher percentages of damaged cells (cTA, 16.4%; V, 13.7%; 1/10 cTA, 9.5%; 1/10 V, 8.6%) (P!.05).

Among cultures exposed to undiluted solutions, cTA damaged more cells than –vTA (P!.01), but the percent- ages were similar for V and cTA (P Z.21). Similarly, among diluted solutions, 1/10 cTA damaged more cells than 1/10 –vTA (P Z.02), but 1/10 V damage was similar to 1/10 cTA damage (P Z.29).

Regarding the dose effect, the percentage of damaged cells was significantly higher in cultures exposed to undi- luted cTA than to 1/10 cTA (P Z .03), and the percentage was also higher in cultures exposed to undiluted V (vehicle alone) than to 1/10 V (P!.01). However, there was no dose response for cultures exposed to vehicle-removed solutions (–vTA and 1/10 –vTA) (P Z.59). These results show a dose- dependent cytotoxic response to solutions containing cTA or its vehicle, but not to TA alone in BSS.

Ultrastructural Changes After 1-Minute Exposures

On TEM, CECs in control cultures had intact cellular membrane, nucleus, and organelles; however, CECs

exposed to cTA, V, or BA 9 mg/mL had various degrees of morphologic damage including significant swelling and disorganization of organelles, disruption of the cell mem- brane, cell lysis, and release of intracellular organelles extracellularly (Figure 3).

Damage to Corneal Endothelial Cells with Longer Exposures

Figure 4 shows the percentages of cells damaged in cultures exposed for 1, 3, 10, or 30 minutes to BSS, V, 1/10 cTA, or 1/10 –vTA. The percentage was similar for cultures exposed to BSS or 1/10 –vTA, even after 30 minutes

Figure 1. Light micrographs of trypsinized rabbit CECs exposed for 1 min-

ute to BSS (control) (A), cTA (B), –vTA (C), V (D), 1/10 cTA (E), 1/10 –vTA (F),

1/10 V (G), or BA 9 mg/mL (H). Dead or dying cells, stained with trypan

blue, vary in size and have bizarre morphology and irregular cell mem-

branes (black arrows; B, D, E, and G). Crystalline deposits are TA particles

(white arrows; B, C, E, and F). Asterisks indicate extensive cell lysis (H) (try-

pan blue stain, original magnification 100).

of exposure (PO.05), whereas exposure to 1/10 cTA for 10 minutes or longer resulted in greater cell damage (10 minutes, P Z .01; 30 minutes, P Z .003). Exposure to V for even 3 minutes caused significant damage (P Z .04, P Z .04, and P Z .03 for 3, 10, and 30 minutes, respectively).

Under LM, trypan blue–positive cells in cultures ex- posed to 1/10 cTA or V for longer periods showed various cell sizes and irregular shapes; some of them appeared bizarrely pleomorphic, swollen, or even lysed (Figure 5).

In general, the longer the cells were exposed to solutions containing the vehicle for cTA, the more severe the morphologic damage.

DISCUSSION

Because corneal endothelium plays a crucial role in maintaining corneal deturgescence, any trauma to this monolayer, such as intracameral injection of a cytotoxic agent, poses the risk for corneal decompensation. To our knowledge, this in vitro study is the first to investigate the risk for CEC damage from contact with the commer- cially available TA suspension used in a variety of ophthal- mic surgical procedures.

Our study found that control and vehicle-removed test solutions had similar lack of effect on CECs, vehicle- containing solutions caused significantly more cell damage than control/vehicle-removed solutions, solutions con- taining vehicle caused similar damage, and the damage caused by vehicle-containing solutions was dose depen- dent, regardless of the concentrations of TA in the solutions. Collectively, these results indicate that the crystalline TA particles did not play a role in cytotoxicity.

Rather, the cytotoxic effects of cTA and 1/10 cTA were mainly attributable to the vehicle. This conjecture was sup- ported by the LM and ultrastructural findings of extensive lysis of CECs exposed to BA, the vehicle in cTA. Moreover, we found that cytotoxicity of the vehicle is time dependent, increasing significantly and progressively with 3-, 10-, and 30-minute exposures to 1/10 cTA or V. These findings imply that during anterior vitrectomy, even a 1/10 dilution of cTA

can cause corneal endothelial damage and that only vehicle-removed solutions should be used.

In an experimental study similar to ours, Hida et al.

found that preservatives in the vehicle for suspension of crystalline cortisone, rather than cortisone itself, could be toxic to the rabbit retina and lens. In addition, preserva- tives, such as benzalkonium chloride, or vehicles in the ir- rigation or other surgical solutions can be implicated in the recent outbreaks of toxic anterior segment inflammation and corneal endothelial destruction after uneventful cata- ract surgery.

Benzyl alcohol is rarely used as the preser- vative in topical ophthalmic solutions because it is irritating, has slow activity, and can dissolve polystyrene.

However, intravenous use of solutions containing BA has caused a fetal

0 5 10 15 20 25

Control cTA -vTA V 1/10 cTA 1/10 -vTA 1/10 V

†

† †

†

Figure 2. Mean percentages (GSD) of trypan blue–

positive (‘‘damaged’’) cultured rabbit CECs after 1-minute exposure to control (BSS), cTA, –vTA, V, 1/10 cTA, 1/10 –vTA, or 1/10 V. Each bar Z 4 cultures (* Z P!.05, Student t test versus control; † Z P!.05, Student t test for 2 groups).

Figure 3. Transmission electron photomicrographs of cultured rabbit CECs after 1-minute exposure to control (BSS) (A; intact cell membrane, nucleus, organelles), cTA, (B), V (C), or BA 9 mg/mL (D). The cTA, V, or BA caused remarkable organelle swelling and disorganization (white arrow; B to D), disruption of cell membrane (black arrow, C), and cell lysis (asterisk; D) (original magnification 10 000).

LABORATORY SCIENCE: CORNEAL ENDOTHELIAL CELLS AND TRIAMCINOLONE TOXICITY

J CATARACT REFRACT SURG - VOL 32, SEPTEMBER 2006

1552

toxic syndrome in premature infants.

Although ocular toxicity of BA has not been proved in humans, these findings support our findings that BA is toxic to CECs.

Even though cTA, V, and 1/10 V (solutions 1, 3 and 6, respectively in Table 1) contained a slightly lower pH (6.1 to 6.8) than BSS, the time of exposure was as short as 1 min- ute. Another study

showed the pH tolerance for rabbit corneal endothelium was between 6.5 and 8.5 with the cor- neas perfused in an in vitro perfusion specular microscope up to 3 hours. We believe that the mildly lower pH in our study does not play a crucial role in CEC cytotoxicity.

The limitation of this study is that we did not perform the experiment in vivo or on excised corneal endothelial cells, which would provide an alternative assessment for such a study. Nevertheless, based on the current model, we evaluated the effect of the dyes

and intracameral anesthetic agents

on corneal endothelial cells, which yielded valid results that are valuable to cataract surgeons. Another lim- itation of this study is that it may overestimate the clinical toxicity of cTA. Toxicity may be less in eyes than in cells in culture because of the actions of proteins and buffered ions in the aqueous humor. Also, TA suspension injected into the anterior chamber for an anterior vitrectomy proce- dure would tend to become diluted by continuous irrigation.

In addition, the use of an ophthalmic viscosurgical device (OVD) tends to protect the corneal endothelium in vivo.

Triamcinolone acetonide suspensions are used exten- sively for ocular conditions because their long-lasting effects can be useful in treating a range of ocular conditions including CME,

central retinal vein occlusion,

diabetic retinopathy,

and choroidal neovascularization.

They are also helpful in visualizing the vitreous during ocular surgeries.

Much of the work in this area used TA 40 mg/mL (Kenalog), which also contains BA 9 mg/mL

as its preservative.

However, the TA suspensions Kenacort-A and Kenalog, which were developed for intra- articular or intramuscular injection, were only recently formulated for ocular off-label use and there are clinical observations that these solutions cause harm.

For exam- ple, residual preservative can cause complications

and residual BA might lead to postoperative sterile or toxic en- dophthalmitis. This hypothesis is supported by the finding that of 520 eyes that had an intravitreal injection of TA,

0 10 20 30 40

1 3 10 30

Pure vehicle 1/10 dilution of TA 1/10 dilution of vehicle- removed TA in BSS BSS

Figure 4. Mean percentages (GSD) of trypan blue–

positive (‘‘damaged’’) cultured rabbit CECs after expo- sure to BSS, 1/10 dilution of –vTA, 1/10 dilution of commercial (V containing) TA, or V for 1 (control), 3, 10, or 30 minutes. Each data point Z mean for 4 cul- tures (* Z P!.05, Student t test versus control).

Figure 5. Light micrographs of rabbit corneal endothelial cells after 30-

minute exposure to control (BSS) (A), 1/10 –vTA in BSS (B), 1/10 dilution

of commercial (V containing) TA (C), and pure V (D). Dead or dying cells

(black arrows; C and D), stained with trypan blue, vary in size and have bi-

zarre morphology and irregular cell membranes. Crystalline deposits are

TA particles (white arrows; B and C) (trypan blue stain, original magnifica-

tion 100).

none that received vehicle-removed TA developed intraoc- ular inflammation.

Thus, we and others urge removal of the vehicle as completely as possible before the TA suspen- sion is used intracamerally.

Three methods to remove the vehicle from cTA have been proposed. Jonas et al.

describe a ‘‘standstill’’ method in which TA suspension is drawn into a syringe that is held vertically for 30 minutes to allow sedimentation of TA crys- tals; the upper layer is ejected, and the settling procedure is repeated twice more before the suspension is used. Because this procedure takes hours, it is not practical when TA sus- pension is needed unexpectedly intraoperatively. Hernaez- Ortega and Soto-Pedre

separated the TA particles from the vehicle by density-gradient centrifugation at 3000 rpm for 5 minutes. This method is simple and rapid, with little loss of TA particles; however, almost no operating rooms are equipped with density-gradient centrifuges. The third method for removing vehicle from cTA, introduced by Burk et al.

and Kumagai,

involves repeatedly passing the TA suspension through Millipore filters and resuspend- ing the TA particles in BSS. Millipore filters are available in most ophthalmic operating rooms, so this method seems the most practical for routine clinical use. It is, however, necessary to titrate the solution to determine the exact concentration,

particularly for intravitreal injection of TA to treat various retinal diseases.

Additional safety measures should be considered when using TA suspensions during ophthalmic surgery. First, even though there was no dose-response for cultures ex- posed to vehicle-removed solutions, we recommend using a 1/10 dilution of –vTA because TA used clinically in ante- rior vitrectomy is its 10-fold dilution rather than its initial commercial concentration (40 mg/mL).

Most surgeons find that a 10-fold dilution of TA suspension provides ex- cellent visualization of prolapsed vitreous. We think the least effective amount of TA should be used as long as visu- alization of vitreous is achieved. Second, even though we found a 1/10 dilution of vehicle-removed TA in BSS to be nontoxic to CECs, we recommend avoiding direct contact of residual vehicle with corneal endothelium, especially in patients with compromised endothelium. To prevent contact, TA should not be injected until after the aqueous humor has been replaced with an air bubble or an OVD.

The intracameral injection should be directed away from the corneal endothelium. Third is whether to leave any TA particles in the eye. Triamcinolone acetonide might reduce the incidence of CME after complicated cataract surgery, but it could lead to prolonged intraocular pressure elevation.

The commercial TA suspension in wide clinical use is cytotoxic to CECs, and the cause of cytotoxicity is the preservative BA in the vehicle, not TA itself. To safely use TA suspensions to visualize the vitreous in complicated

anterior segment surgeries, the vehicle in the TA suspension should be removed. A 1/10 dilution of vehicle-removed TA appears safe for cultured rabbit CECs, and we recommend this preparation for intracameral injection into human eyes. We also recommend removing the TA as completely and quickly as possible. However, more clinical studies are needed to clarify the long-term safety of TA suspensions on human corneal endothelium and the retina.

REFERENCES

1. Ng DT, Rowe NA, Francis IC, et al. Intraoperative complications of 1000 phacoemulsification procedures: a prospective study. J Cataract Re- fract Surg 1998; 24:1390–1395

2. Berrod JP, Sautiere B, Rozot P, Raspiller A. Retinal detachment after cataract surgery. Int Ophthalmol 1996-97; 20:301–308

3. Burk SE, Da Mata AP, Snyder ME, et al. Visualizing vitreous using Kena- log suspension. J Cataract Refract Surg 2003; 29:645–651

4. Peyman GA, Cheema R, Conway MD, Fang T. Triamcinolone acetonide as an aid to visualization of the vitreous and the posterior hyaloid dur- ing pars plana vitrectomy. Retina 2000; 20:554–555

5. Kumagai K. Introduction of a new method for the preparation of tri- amcinolone acetonide solution as an aid to visualization of the vitre- ous and the posterior hyaloid during pars plana vitrectomy. Retina 2003; 23:881–882

6. Hida T, Chandler D, Arena JE, Machemer R. Experimental and clinical observations of the intraocular toxicity of commercial corticosteroid preparations. Am J Ophthalmol 1986; 101:190–195

7. Roth DB, Chieh J, Spirn MJ, et al. Noninfectious endophthalmitis asso- ciated with intravitreal triamcinolone injection. Arch Ophthalmol 2003; 121:1279–1282

8. Sutter FKP, Gilles MC. Pseudo-endophthalmitis after intravitreal injec- tion of triamcinolone. Br J Ophthalmol 2003; 87:972–974

9. Nelson ML, Tennant MTS, Sivalingam A, et al. Infectious and presumed noninfectious endophthalmitis after intravitreal triamcinolone aceto- nide injection. Retina 2003; 23:686–691

10. Chang Y-S, Tseng S-Y, Tseng S-H, et al. Comparison of dyes for cataract surgery. Part I: cytotoxicity to corneal endothelial cells in a rabbit model. J Cataract Refract Surg 2005; 31:792–798

11. Chang Y-S, Tseng S-Y, Tseng S-H. Cytotoxicity of lidocaine or bupiva- caine on corneal endothelial cells in a rabbit model. Cornea 2006;

25:290–296

12. Hernaez-Ortega MC, Soto-Pedre E. A simple and rapid method for pu- rification of triamcinolone acetonide suspension for intravitreal injec- tion. Ophthalmic Surg Lasers Imaging 2004; 35:350–351

13. Monson MC, Mamalis N, Olson RJ. Toxic anterior segment inflamma- tion following cataract surgery. J Cataract Refract Surg 1992; 18:

184–189

14. Liu H, Routley I, Teichmann KD. Toxic endothelial cell destruction from intraocular benzalkonium chloride. J Cataract Refract Surg 2001; 27:

1746–1750

15. Mullen W, Shepherd W, Labovitz J. Ophthalmic preservatives and ve- hicles. Surv Ophthalmol 1973; 17:469–483

16. Brown WJ, Buist NRM, Cory Gipson HT, et al. Fatal benzyl alcohol poi- soning in a neonatal intensive care unit [letter]. Lancet 1982; 319:1250 17. Gonnering R, Edelhauser HF, Van Horn DL, Durant W. The pH toler- ance of rabbit and human corneal endothelium. Invest Ophthalmol Vis Sci 1979; 18:373–390

18. Conway MD, Canakis C, Livir-Rallatos C, Peyman GA. Intravitreal triam- cinolone acetonide for refractory chronic pseudophakic cystoid mac- ular edema. J Cataract Refract Surg 2003; 29:27–33

LABORATORY SCIENCE: CORNEAL ENDOTHELIAL CELLS AND TRIAMCINOLONE TOXICITY

J CATARACT REFRACT SURG - VOL 32, SEPTEMBER 2006

1554

19. Greenberg PB, Martidis A, Rogers AH, et al. Intravitreal triamcinolone acetonide for macular oedema due to central retinal vein occlusion [letter]. Br J Ophthalmol 2002; 86:247–248

20. Jonas JB, Degenring RF, Kamppeter BA, et al. Duration of the effect of intravitreal triamcinolone acetonide as treatment for diffuse diabetic macular edema. Am J Ophthalmol 2004; 138:158–160

21. Jonas JB, Hayler JK, So¨fker A, Panda-Jones S. Intravitreal injection of crystalline cortisone as adjunctive treatment of proliferative diabetic retinopathy. Am J Ophthalmol 2001; 131:468–471

22. Ranson NT, Danis RP, Ciulla TA, Pratt L. Intravitreal triamcinolone in subfoveal recurrence of choroidal neovascularisation after laser

treatment in macular degeneration [correspondence]. Br J Ophthal- mol 2002; 86:527–529

23. Yamakiri K, Uchino E, Kimura K, Sakamoto T. Intracameral triamcino- lone helps to visualize and remove the vitreous body in anterior chamber in cataract surgery. Am J Ophthalmol 2004; 138:650–652 24. Jonas JB, Kreissig I, Degenring RF. Endophthalmitis after intravitreal

injection of triamcinolone acetonide [letter]. Arch Ophthalmol 2003;

121:1663–1664

25. Spandau UHM, Derse M, Schmitz-Valckenberg P, et al. Dosage depen-

dency of intravitreal triamcinolone acetonide as treatment for dia-

betic macular oedema. Br J Ophthalmol 2005; 89:999–1003