A Comparison of Preservative-Free Diclofenac and Preserved Diclofenac Eye Drops after Cataract Surgery in Patients with Diabetic Retinopathy

Abstract

Purpose:

The aim of this study was to compare the anti-inflammatory efficacy of preservative-free and preserved 0.1% diclofenac eye drops for the management of postoperative inflammation after cataract surgery in patients with nonproliferative diabetic retinopathy and in normal controls.

Methods:

Forty-two diabetic patients and 50 normal control patients who underwent small-incision phacoemulsification cataract surgery bilaterally received topical preservative-free diclofenac in 1 eye and preserved diclofenac in the other eye. The corrected distance visual acuity (CDVA) as determined by a logarithm of the minimum angle of resolution (logMAR) chart, intraocular pressure (IOP), foveal thickness (FT) using optical coherence tomography (OCT), and the anterior chamber flare (ACF) score measured with a laser flare cell meter were monitored for 12 weeks after surgery.

Results:

In the eyes with diabetic retinopathy, there were no significant differences in CDVA, IOP, FT, and ACF score between the right and left eyes at the initial exam. After cataract surgery, changes in CDVA, IOP, and FT were not influenced by the preservative in the diclofenac eye drops. In contrast, the ACF score in the eyes treated with preserved diclofenac showed slower recovery from postoperative inflammation than the eyes treated with preservative-free diclofenac. In the normal control eyes, similar but milder changes were observed in each of the clinical parameters.

Conclusions:

Because preservative suppressed the anti-inflammatory efficacy of topical diclofenac after cataract surgery, preservative-free diclofenac may have an improved safety profile during postoperative treatment, especially in patients with diabetic retinopathy.

Introduction

Topical diclofenac sodium is a potent nonsteroidal anti-inflammatory drug (NSAID) that has widespread use worldwide as an anti-inflammatory and analgesic agent for ocular inflammatory conditions, especially after cataract surgery. Its anti-inflammatory activity results mainly from cyclooxygenase (COX) inhibition, leading to a reduction in prostaglandin synthesis. Compared to anti-inflammatory corticosteroid eye drops, NSAID eye drops do not cause a postoperative increase in intraocular pressure (IOP)¹ or cystoid macular edema (CME).²

Most eye drops contain preservatives to provide a level of antimicrobial activity that limits secondary bacterial, mycotic, and amoebal ocular infections caused by contaminated solutions, and to prolong the half-life of the drugs by preventing biodegradation and maintaining drug potency. Common topical ophthalmic solution preservatives are reported to have the potential to harm the eye, and toxicity from pharmaceutical agents can result in decreased visual acuity and/or patient comfort.³ Although a recent study revealed that preservative-free diclofenac eye drops exhibited a significantly better postoperative subjective and objective tolerability when compared with preserved diclofenac,⁴ another study concluded that preservative suppression did not alter diclofenac efficacy after cataract surgery. However, neither study was a case-controlled study conducted in patients with eligible eyes, where 1 eye was treated with preservative-free diclofenac and the other eye with preserved diclofenac.

In this prospective case-controlled study, we compared and evaluated eyes treated with topical preservative-free diclofenac and eyes treated with preserved diclofenac for postoperative inflammation after cataract surgery. Eligible patients had diabetes mellitus, and patients without diabetes were used as normal controls. We studied patients with diabetes because eyes with diabetic retinopathy have characteristic inflammation that often induces postoperative macular thickening,^5,6 which requires more careful observation during the postoperative period.⁷

Methods

Patient eligibility

In this prospective study, diabetic patients with simple or nondiabetic retinopathy and normal control patients requiring cataract surgery in bilateral eyes were recruited. The exclusion criteria in each group were: (1) The difference of corrected distance visual acuity (CDVA) with logarithm of the minimum angle of resolution (logMAR) chart between both eyes of each patient was more than 0.2; (2) the eye with the better CDVA was less than 0.5; (3) the difference in refractive error between both eyes of each patient was more than 2.0; (4) the difference in averaged foveal thickness (FT) between both eyes of each patient was more than 50 μm; and (5) patients with a history of any ocular disease other than refractive errors and nonproliferative diabetic retinopathy.

This study was approved by the Ethics Committee for Clinical Research at NTT East Japan Tohoku Hospital. After informing the patient of the purpose of this study and the possible outcomes, informed consent was obtained from all patients prior to the intervention. Using procedures that conformed to the tenets of the Declaration of Helsinki, all patients received a comprehensive ocular examination before and after the treatments. The presence of diabetic retinopathy was established using direct and indirect ophthalmoscopy and slit-lamp biomicroscopy of the posterior segment with a Volk Superfield NC lens (Volk, Mentor, OH). Fundus photographs were taken at the follow-up examinations. Diabetic patients were monitored and treated appropriately by their internists for their general diabetic and hypertensive conditions.

Drugs

Before cataract surgery, the topical antimicrobial drug 0.5% levofloxacin ophthalmic solution (Clavit^®; Santen Pharmaceutical Co. Ltd, Osaka, Japan), which has no preservative, was administered 4 times/day in both eyes of each patient. The study medications were preservative-free 0.1% diclofenac sodium eye drops (Diclostar PF^®, Nihon Tenganyaku Kennkyusho Co. Ltd., Nagoya, Japan), or preserved 0.1% diclofenac sodium eye drops containing the preservative chlorobutanol (Diclod^®; Wakamoto Pharmaceutical Co. Ltd., Tokyo, Japan).

Study design

All operations were performed in a standard way by the same experienced surgeon (K.Y.). The surgical technique used in this study consisted of a continuous curvilinear capsulorrhexis (CCC), phacoemulsification, and implantation of a foldable acrylic intraocular lens (IOL; Acryfold^® PY-60AD; HOYA Medicals, Tokyo, Japan) into the lens capsule through a 2.5-mm corneal incision. The incision remained sutureless. The surgery took less than 10 min, and no complications occurred. In each patient, the eye with the lower visual acuity was operated first, followed by the second eye within 1 week. If visual acuities in both eyes were the same, then the right eye was chosen first.

After the next-day examination (defined as day 1), topical administration of an anti-inflammatory drug was started; Diclod^® and Clavid^® were applied topically 4 times/day in the right eyes, and Diclostar^® and Clavid^® were applied topically 4 times/day in the left eyes. Drug applications were continued for at least 12 weeks after surgery.

CDVA using the logMAR chart (5 m) (NEITZ LVC-10, Tokyo, Japan), IOP measured with an applanation tonometer, FT obtained from optical coherence tomography (OCT) images (Cirrus HD-OCT^®, Humphrey Zeiss Inc, San Leandro, CA), and anterior chamber flare (ACF) score assessed by a Laser Flare and Cell Meter (LFCM 1000; Kowa, Tokyo, Japan) were measured preoperatively and postoperatively at day 1 and at 1, 4, 8, and 12 weeks following cataract surgery.

FT was calculated by 2 experienced examiners by manually averaging the foveal thickness of the horizontal and vertical images using the caliper tool built into the OCT software. To obtain the ACF score, 7 consecutive laser flare readings were measured, the highest and lowest readings were discarded, and then the remaining 5 readings were averaged. ACF values were expressed in photons per millisecond (ph/ms). Calibration of the Laser Flare and Cell Meter was performed on a regular basis.

Statistical analysis

The data are presented as mean±standard deviation. Statistical differences between follow-up points during the postoperative clinical course of the same eyes were assessed using the Wilcoxon signed-rank test, and the differences between the right and left eyes were assessed using the Mann Whitney U-test (SPSS, Chicago, IL). In all cases, p<0.05 was considered to be significant.

Results

Forty-two patients with type 2 diabetes (84 eyes) participated in this study; none of these patients had any prior history of retinal photocoagulation. Twenty-three of the 42 patients underwent cataract surgery in the right eye first and then 1 week later in the left eye, whereas the remaining 19 patients underwent cataract surgery first in the left eye and then in the right eye. The ages of the diabetic patients ranged from 53 to 88 years, with a mean of 72.6±9.8 years, and the duration of their diabetes ranged from 3 to 25 years with a mean of 10.3±5.3 years. At their initial examination, the average glycolsylated hemoglobin (HbA1c) was 6.12±0.30%, ranging from 5.4% to 6.8%.

Fifty normal control patients (100 eyes) participated in this study. Their ages ranged from 52 to 89 years with a mean of 73.9±9.8 years. Twenty-eight of the 50 patients underwent cataract surgery in the right eye first and then 1 week later in the left eye, whereas the remaining 22 patients underwent cataract surgery first in the left eye and then in the right eye. There was no statistically significant difference in age between the diabetic group and the control group, which defines the normal control group as an age-matched control group.

In this prospective study, topical anti-inflammatory NSAIDs were applied after the examination on postoperative day 1. Therefore, the clinical parameters CDVA, IOP, FT, and ACF at postoperative day 1 were not influenced by the topical NSAIDs, but only by cataract surgery. The changes seen in the clinical parameters from day 1 through 12 weeks postoperatively were influenced by the topical NSAIDs.

Alterations in clinical parameters before topical NSAID application

The initial evaluation of logMAR CDVA revealed no statistically significant difference among the 4 groups, with the right eyes measuring 0.85±0.20 and the left eyes measuring 0.86±0.21, both in the patients with diabetic retinopathy and in the patients in the normal control groups. After cataract surgery, the CDVAs in all 4 groups were improved. It was noted that CDVA at postoperative day 1 was statistically significantly worse in the diabetic eyes than in the control eyes (Fig. 1A).

FIG. 1.

Changes in the clinical parameters logarithm of the minimum angle of resolution (logMAR) visual acuity (A), intraocular pressure (IOP) (B), foveal thickness (FT) (C), and anterior chamber flare (ACF) score (D) preoperatively and on postoperative day 1. The black circles indicate right eyes and the white circles indicate left eyes of the patients with diabetic retinopathy. Similarly, the dark gray circles and light gray circles indicate the right and left eyes of the control patients. Note that none of the eyes have yet received diclofenac treatment. The longitudinal bar in each circle indicates standard deviation. An asterisk (*) indicates statistical significance (p<0.05). CDVA, Corrected distance visual acuity.

Similarly, the initial evaluation of IOP revealed no statistically significant difference among the 4 groups. The right eyes measured 16.2±2.21 mmHg and the left eyes measured 15.9±2.08 mmHg in the patients with diabetic retinopathy, and right and left eyes in the normal control groups measured (16.0±2.17 mmHg). On postoperative day 1, the IOP in each of the 4 groups had not changed (Fig. 1B).

The initial evaluation of FT revealed no statistically significant difference between the right eyes (183.4±18.7 μm) and the left eyes (182.6±17.9 μm) in the patients with diabetic retinopathy. Also, there was no statistically significant difference in FT between the right (169.4±11.6 μm) and left eyes (166.5±11.4 μm) in the control groups. However, the initial FT in the groups with diabetic retinopathy was statistically significantly thicker than that in the control groups. At postoperative day 1 following cataract surgery, although the FT in each group had not changed, the FT in the eyes with diabetic retinopathy was still statistically significantly thicker than in the control eyes (Fig. 1C).

At the initial assessment of the ACF score, there was no statistically significant difference between the right eyes (13.4±6.7 ph/ms) and the left eyes (13.7±8.6 ph/ms) of the patients with diabetic retinopathy. There was also no statistically significant difference between the right (9.8±4.7 ph/ms) and left (9.4±5.3 ph/ms) eyes in the control groups. However, the initial ACF scores in the eyes with diabetic retinopathy were statistically significantly higher than in the eyes in the control groups. At postoperative day 1 following cataract surgery, although the ACF scores in both diabetic retinopathy and control eyes showed a statistically significant increase, the ACF score in the eyes with diabetic retinopathy was still statistically significant higher than in the eyes in the control groups (Fig. 1D).

Postoperative changes in clinical parameters after application of topical NSAIDs

The postoperative clinical course of CDVA is depicted in Fig. 2. The patients with diabetic retinopathy showed a statistically significant improvement in CDVA in both eyes from postoperative day 1 to week 1. There was no statistically significant difference between the right (preserved diclofenac-treated) eyes and left (preservative-free diclofenac-treated) eyes during the postoperative course (Fig. 2A). Similarly, in the patients with the normal control eyes, there was no statistically significant difference in CDVA between the two eyes in the follow-up period in discussion (Fig. 2B).

FIG. 2.

Alterations in postoperative corrected distance visual acuity (CDVA) with logarithm of the minimum angle of resolution (logMAR) chart (A) In patients with diabetic retinopathy, the black circles indicate right eyes treated with preserved diclofenac and the white circles indicate left eyes treated with preservative-free diclofenac. In both groups of diabetic eyes, CDVA improved with time, and there was no significant difference between the right and left eyes during the postoperative clinical course. (B) In the control patients, the dark gray and light gray circles indicate right (preserved diclofenac) and left (preservative-free diclofenac) eyes. Alterations in CDVAs were similar to those in the diabetic eyes. The longitudinal bar in each circle indicates standard deviation. An asterisk (*) indicates statistical significance (p<0.05).

The postoperative clinical course of IOP is depicted in Fig. 3. Both the patients with diabetic retinopathy and the normal controls had a gradual decrease over time in the IOP in both eyes, and the decrease was statistically significant from postoperative week 4 to week 12 (Fig. 3A,B).

FIG. 3.

Alterations in postoperative intraocular pressure (IOP) (A) In patients with diabetic retinopathy, the black circles indicate right eyes treated with preserved diclofenac, and the white circles indicate left eyes treated with preservative-free diclofenac. In both groups of diabetic eyes, IOP decreased with time up to postoperative week 4 and stabilized after that, and there was no significant difference between the right and left eyes during the postoperative clinical course. (B) In the control patients, the dark gray and light gray circles indicate right (preserved diclofenac) and left (preservative-free diclofenac) eyes. Alterations in IOPs were similar to those in the diabetic eyes. The longitudinal bar in each circle indicates standard deviation. An asterisk (*) indicates statistical significance (p<0.05).

The postoperative clinical course of FT is depicted in Fig. 4. In the eyes with diabetic retinopathy, the FTs in both the right and left eyes gradually increased and showed a statistically significant increase by postoperative week 4. However, there was no statistically significant difference in FT between the two eyes at any point in time (Fig. 4A). The FTs in the right and left control eyes were not altered during the postoperative clinical course, and there was no statistically significant difference in FT at any point in time between the two eyes (Fig. 4B).

FIG. 4.

Alterations in postoperative foveal thickness (FT). (A) In patients with diabetic retinopathy, the black circles indicate right eyes treated with preserved diclofenac, and the white circles indicate left eyes treated with preservative-free diclofenac. In both groups of diabetic eyes, FT increased with time up to postoperative week 4, and there was no significant difference between the right and left eyes during the postoperative clinical course. (B) In the control patients, the dark gray and light gray circles indicate right (preserved diclofenac) and left (preservative-free diclofenac) eyes. During the postoperative clinical course, the FT in both groups of control eyes was not changed. The longitudinal bar in each circle indicates standard deviation. An asterisk (*) indicates statistical significance (p<0.05).

The postoperative ACF scores are depicted in Fig. 5. In the eyes with diabetic retinopathy, the ACF scores in both eyes were decreased (improved) with time during the postoperative clinical course. Compared with the ACF score on postoperative day 1, the ACF scores at postoperative week 1 were statistically significantly decreased in both eyes. Although the ACF scores in the left (preservative-free diclofenac-treated) eyes were decreased more significantly than those in the right (preserved diclofenac-treated) eyes at postoperative weeks 1 and 4, there was no significant difference in ACF scores between either group of eyes at weeks 8 and 12 (Fig. 5A). In the control eyes, the ACF scores in both eyes decreased with time during the postoperative clinical course. Only at postoperative week 1 did the ACF scores in the left (preservative-free diclofenac-treated) eyes show a significant decrease as compared to the right (preserved diclofenac-treated) eyes. However, there was no significant difference in ACF scores between the right and left eyes at any other postoperative periods (Fig. 5B).

FIG. 5.

Alterations in postoperative anterior chamber flare (ACF) score. (A) In patients with diabetic retinopathy, the black circles indicate right eyes treated with preserved diclofenac, and the white circles indicate left eyes treated with preservative-free diclofenac. In both groups of diabetic eyes, the ACF score decreased with time, but in the preservative treated eyes it was statistically significantly higher than in the preservative-free eyes at postoperative weeks 1 and 4. (B) In the control patients, dark gray and light gray circles indicate right (preserved diclofenac) and left (preservative-free diclofenac) eyes. In both groups of control eyes, ACF scores decreased with time, but those in the preservative treated eyes were statistically significantly higher than those in the preservative-free eyes only at postoperative week 1. The longitudinal bar in each circle indicates standard deviation. An asterisk (*) indicates statistical significance (p<0.05).

Discussion

According to this prospective case–control study, although preserved topical diclofenac did not affect the postoperative changes in CDVA, IOP, and FT after cataract surgery, topical preservative did alter the anti-inflammatory effect of topical NSAIDs during the early postoperative period. The interruption of the anti-inflammatory effect by preservative was especially observed in eyes with diabetic retinopathy, as evidenced by higher ACF scores.

Although advances in cataract surgery have resulted in improved postoperative outcomes with less intraocular inflammation and faster visual recovery, noncontrolled inflammation can lead to complications, such as cystoid macular edema. Thus, topical NSAIDs are used routinely after cataract surgery. Diclofenac has the unique quality of inhibiting both the cyclooxygenase and the lipoxygenase pathways,^8–10 which provides in turn a spectrum of activity more similar to corticosteroids, leading us to expect an increased anti-inflammatory effect without a steroid-induced increase in IOP.

Nearly all eye drops contain preservatives to decrease contamination, but these preservatives can add to the toxicity of eye drops and cause ocular surface inflammation.³ A previous study revealed that there was no difference in the efficacy and safety in controlling postoperative inflammation when topical diclofenac with and without benzalkonium chloride (BAK) was used.¹¹ Although commercial topical diclofenac (Diclod^®) contains chlorobutanol, which has lower toxicity than BAK,¹² our data suggest that the anti-inflammatory effect of topical diclofenac was significantly suppressed by the preservative chlorobutanol. This discrepancy may be due to the difference in the study design. In the previous study, 194 normal patients were divided into 2 groups at random, and a comparison between these 2 groups of patients was performed. In contrast, in our study, 42 patients with diabetic retinopathy and 50 normal patients underwent bilateral cataract surgery, and a comparison between the right and left eyes of the same patient was investigated. The selection of subjects after bilateral cataract surgery can minimize the bias introduced by the systemic conditions.

Chlorobutanol, the detergent preservative used in this study, has a broad spectrum of antimicrobial action. It works by disorganizing the lipid structure of the cell membrane, which increases the permeability of the cell, thus leading to cell lysis,³ causing cell retraction and cessation of normal cytokinesis, cell movement, and mitotic activity. It disrupts the barrier and transport properties of the corneal epithelium as well as inhibiting the utilization of oxygen by the cornea.¹² Thus, topical chlorobutanol can cause inflammation of ocular tissue. Although the exact mechanism by which chlorobutanol inhibits the anti-inflammatory effect of diclofenac remains unclear, chlorobutanol may accelerate postoperative inflammation and thus compete with the anti-inflammatory effect of diclofenac.

The intraocular condition in diabetic patients is unique, and even with preclinical retinopathy, a breakdown of the blood–retinal barrier is observed.¹³ Furthermore, vitreous levels of hemodynamic and/or inflammatory-related cytokines are reported to be elevated.^14–16 Thus, diabetic retinopathy is characterized by inflammation. Even before cataract surgery in this study, compared with normal controls, both FT and ACF in the eyes with diabetic retinopathy were statistically significantly higher (Fig. 1). However, a postoperative increase in FT and ACF, which can lead to CME, was prominent in patients with diabetic retinopathy. Fortunately, in our study, definite severe postoperative CME with visual dysfunction was not observed. Even in diabetic eyes, the clinical incidence of postoperative CME is reported to be 0% to 6%, and the incidence of angiographic signs for CME is up to 55%.¹⁷ Recent studies using OCT showed macular thickening after cataract surgery, even with postoperative treatment with topical diclofenac, and the thickening is more prominent in patients with diabetic retinopathy,^5,6 which is similar to our results.

Thus, in patients with diabetes, postoperative inflammation can be more severe. Even diclofenac, which has a strong anti-inflammatory effect, may not fully suppress either postoperative- or preservative-induced inflammation. Recently, 0.1% diclofenac sodium eye drops have become commercially available as 2 formulations, 1 with and 1 without preservative. Because topical preservative-free diclofenac can avoid preservative-induced inflammation, it should be considered for suppression of postoperative inflammation following cataract surgery in patients with diabetic retinopathy. Of course, we must point out that preservative-free eye drops contain no antimicrobial activity, and thus we need to assure that the medicine is stored properly.

Footnotes

Acknowledgments

The authors thank Dr. Norio Sugimoto in Theranostic Instruments Research Laboratories for his technical support of statistical analyses and useful comments. This study was supported by a 2010 Scientific Grant in Aid for Clinical Research in NTT Hospital.

Author Disclosure Statement

The authors have no competing financial interests in this work.

References

Strelow

S.A.

, Sherwood

M.B.

, Broncato

L.J.

, Napler

, Driebe

W.T.J.

, Guy

J.R.

, Vickers

F.F.

, Mellars

Effect of diclofenac sopdium ophthalmic solution on intraocular pressure following cataract extraction. Ophthalmic. Surg., 1992; 23:170–175.

Miyake

, Masuda

, Shirato

, Oshika

, Eguchi

, Hoshi

, Majima

, Kimura

, Hayashi

Comparison of diclofenac and fluorometholone in preventing cystoid macular edema after small incision cataract surgery: A multicentered prospective trial. Jpn. J. Ophthalmol., 2000; 44:58–67.

Noecker

Effects of common ophthalmic preservatives on ocular health. Adv. Ther., 2001; 18:205–215.

Maca

S.M.

, Amon

, Findl

, Kahraman

, Barisani-Asenbauer

Efficacy and tolerability of preservative-free and preserved diclofenac and preserved ketrolac eyedrops after cataract surgery. Am. J. Ophthalmol., 2010; 149:777–784.

Endo

, Kato

, Haruyama

, Shoji

, Kitano

Efficacy of bromfenac sodium ophthalmic solution in preventing cystoid macular oedema after cataract surgery in patients with diabetes. Acta Ophthalmol., 2009; 88:896–900.

Shimura

, Nakazawa

, Yasuda

, Nishida

Diclofenac prevents an ealry event of macular thickening after cataract surgery in patients with diabetes. J. Ocul. Pharmacol. Ther., 2007; 23:284–291.

Kern

T.S.

Contributions of inflammatory processes to the development of the early stages of diabetic retinopathy. Exp. Diabetes Res., 2007; 2007:95103.

Flach

A.J.

Topical nonstreoidal anti-inflammatory drugs in ophthalmology. Int. Ophthalmol. Clin., 2002; 42:1–11.

Gaynes

B.I.

, Fiscella

Topical nonsteroidal anti-inflammatory drugs for ophthalmic use: A safety review. Drug Saf., 2002; 25:233–250.

10.

Schalnus

Topical nonsteroidal anti-inflammatory therapy in ophthalmology. Ophthalmologica, 2003; 217:89–98.

11.

Bodaghi

, Weber

M.E.

, Arnoux

Y.V.

, Jaulerry

S.D.

, Hoang

P.L.

, Colin

Comparison of the efficacy and safety of two formulations of diclofenac sodium 0.1% eyedrops in controlling postoperative inflammation after cataract surgery. Eur. J. Ophthalmol., 2005; 15:702–711.

12.

Epstein

S.P.

, Ahdoot

, Marcus

, Asbell

P.A.

Comparative toxicity of preservatives on immortalized corneal and conjunctival epithelial cells. J. Ocul. Pharmacol. Ther., 2009; 25:113–119.

13.

Lobo

C.L.

, Bernardes

R.C.

, Cunha-Vaz

J.G.

Alterations of the blood-retinal barrier and retinal thickness in preclinical retinopathy in subjects with type 2 diabetes. Arch. Ophthalmol., 2000; 118:1364–1369.

14.

Funatsu

, Yamashita

, Ikeda

, Mimura

, Eguchi

, Hori

Vitreous levels of interleukin-6 and vascular endothelial growth factor are related to diabetic macular edema. Ophthalmology, 2003; 110:1690–1696.

15.

Funatsu

, Yamashita

, Sakata

, Noma

, Mimura

, Suzuki

, Eguchi

, Hori

Vitreous levels of vascular endothelial growth factor and intracellular adhesion molecule 1 are related to diabetic macular edema. Ophthalmology, 2005; 112:806–816.

16.

Petel

J.I.

, Saleh

G.M.

, Hykin

P.G.

, Gregor

A.J.

, Cree

I.A.

Concentration of haemodynamic and inflammatory related cytokines in diabetic retinopathy. Eye, 2008; 22:223–228.

17.

Yavas

G.F

, Ozturk

, Kusbeci

Preoperative topical indomethacin to prevent pseudophakic cystoid macular edema. J. Cataract Refract. Surg., 2007; 33:804–807.