longitudinal effect of topical antiglaucoma medications on central corneal thickness

Original Article

Longitudinal effect of topical antiglaucomamedications on central corneal thicknessDeepa Viswanathan FRCS,1 Ivan Goldberg FRANZCO2 and Stuart L Graham FRANZCO PhD1,2

1Australian School of Advanced Medicine, Macquarie University and 2Department of Ophthalmology, Save Sight Institute, University ofSydney, Sydney, New South Wales, Australia

ABSTRACT

Background: To determine the change in centralcorneal thickness over time and whether the use oflong-term topical antiglaucoma medications influ-ences central corneal thickness.

Design: Case control study with retrospective andprospective data collection.

Participants: One hundred eighty-seven eyes of 187glaucoma patients (mean follow up 6.92 � 1.67years) being treated with topical antiglaucoma medi-cations (at least 3 years) with no history of surgery orlaser were included and compared with 100 eyes of100 age–matched, untreated control subjects (meanfollow up 6.58 � 1.93 years) who were glaucomasuspects with normal intraocular pressure not onany treatment.

Methods: Demographic data, central corneal thick-ness and intraocular pressure were collected at initialglaucoma diagnosis and at most recent visit, andfindings were compared between two groups.

Main Outcome Measures: Mean change in centralcorneal thickness in microns (mm).

Results: Central corneal thickness fell significantly(P < 0.0001) in treated eyes but not in control eyes(P = 0.18); mean central corneal thickness reduc-tion was 12.29 � 13.65 mm in treated eyes and1.17 � 8.75 mm in controls. Among treated eyes,central corneal thickness reduction was significant

(P < 0.0001) in those treated with either prostaglan-dins or a combination of prostaglandin and beta-blockers, while no significant reduction occurred ineyes treated with only beta-blockers (P = 0.15) whencompared with control eyes.

Conclusions: Prostaglandins appear to be associatedwith a small but significant central corneal thicknessreduction over time. Serial central corneal thicknessmeasurements might be helpful in glaucoma pati-ents, particularly those on prostaglandins.

Key words: central corneal thickness, antiglaucomamedications, prostaglandins, betablockers, intraocularpressure.

INTRODUCTION

Because of its influence on the measurement accuracyof intraocular pressure (IOP) by applanation tonom-etry, central corneal thickness (CCT) measurementhas assumed importance in the management of glau-coma patients.1–4 In addition, the Ocular Hyperten-sion Treatment Study (OHTS) demonstrated thatCCT is a powerful predictive factor for the develop-ment of primary open-angle glaucoma (POAG) fromocular hypertension.5

Earlier cross-sectional studies suggested that CCTmay decrease over time;6–8 this was also corroboratedby two longitudinal studies, one by Weizer et al. on39 subjects9 and the other by Brandt et al.10 on OHTSparticipants who demonstrated a reduction in CCTcomparable with other cross-sectional studies. Theuse of topical antiglaucoma medications could be

� Correspondence: Dr Deepa Viswanathan, Hub 1, 2 Technology Place, Australian School of Advanced Medicine, Macquarie University, Sydney, NSW

2109, Australia. Email: [email protected]

Received 20 April 2012; accepted 17 August 2012.

Competing/conflicts of interest: No stated conflict of interest.

Funding sources: No stated funding sources.

bs_bs_banner

Clinical and Experimental Ophthalmology 2013; 41: 348–354 doi: 10.1111/j.1442-9071.2012.02870.x

© 2012 The AuthorsClinical and Experimental Ophthalmology © 2012 Royal Australian and New Zealand College of Ophthalmologists

associated with changes in the corneal structure,11,12

and prostaglandin analogues may decrease CCTvalues.13–18 However, in one recent study, prostaglan-din analogues were found to increase the CCT valueslightly.19

As the maximal mean follow up for earlier studiesspecifically evaluating the effect of topical medica-tions on CCT was less than 48 months, there is aneed for investigating both normal and glaucoma-treated populations with longer follow up. We,therefore, examined whether there are demonstrableCCT changes over time in treated glaucoma subjectscompared with non-treated subjects, and comparedthe long-term effects of various types of topicalantiglaucoma medications on CCT.

METHODS

A total of 187 eyes of 187 patients (one eye perpatient) with glaucoma being treated with topicalantiglaucoma medications were included and werecompared with 100 eyes of 100 glaucoma suspects(one eye per subject) who served as age-matchedcontrols. The study was conducted in a tertiary refer-ral glaucoma practice, following the tenets of theDeclaration of Helsinki after obtaining approval fromthe Institutional Review Board/Ethics committee.

The inclusion criteria were patients with newlydiagnosed glaucoma, either POAG or normal tensionglaucoma (NTG), who had not received topical glau-comatous treatment prior to their initial CCT meas-urement. Patients must have had treatment for atleast 3 years with a particular class of drugs. Whenmore than one class of drugs were used in a patient,they were included as a combination group if eachclass of drugs was used for at least 3 years duringfollow up. The inclusion criteria for the ‘control’group were patients who were glaucoma suspects,based on suspicious but not yet glaucomatous discswith normal visual fields and IOP, and who did notrequire treatment. Some subjects had only a familyhistory of glaucoma but were being checked periodi-cally. Patients with corneal disease including dryeye, past history of ocular trauma, any intraocularsurgery, laser treatment, recent contact lens wear anddiabetes on drug treatment were excluded. Therequirements for at least 3 years of consistent druguse and no laser or surgery substantially reduced thesample size available for this study but removedconfounders that might affect long-term CCT.

For the purposes of analysis only, the glaucomasubjects were divided into POAG and NTG sub-groups. POAG was defined as a typical glaucoma-tous optic nerve with focal or diffuse structuraldamage with or without visual field defect(s) dem-onstrated in at least two consecutive reliable exami-nations (Glaucoma Hemifield Test abnormal) and an

IOP equal to or greater than 21 mmHg (Goldmannapplanation tonometry). No subjects in this studyhad a visual field defect without some associated discchange. NTG was defined as for POAG, but with anIOP never recorded as greater than 20 mmHg.

At the initial visit, age, gender, race and familyhistory of glaucoma in first-degree relatives wererecorded per patient. CCT measurements were per-formed before IOP measurements after instillingtopical anaesthetic eye drops by experienced orthop-tists using an ultrasound pachymeter recording amean of five consecutive readings. The pachymeterunderwent an upgrade in 2005; therefore, CCT meas-urements taken prior to 2005 were performed usingTomey AL-2000 (Tomey Inc, Nagoya, Japan), andthose after 2005 were performed using a newerversion, the Tomey SP-3000 (Tomey Inc, Nagoya,Japan). At the subject’s most recent visit, CCT andIOP were again recorded, and their history of topicaltherapy was recorded. CCT measurements at base-line before treatment and after treatment were ana-lyzed (see succeeding sections). Thus, each patient’sbaseline CCT and IOP data were collected retrospec-tively, and CCT and IOP data at present visit werecollected prospectively.

Statistics

Statistical analysis was performed using SPSS forWindows version 19 (SPSS, Inc., Chicago, IL, USA).The independent samples t-test was used to comparecharacteristics including age, baseline IOP and base-line CCT between the treated and untreated eyes.The changes in CCT values over time withinthe different subgroups were evaluated using apaired sample t-test. Fisher’s test was performedto evaluate the gender difference between thetwo groups. A P-value of �0.05 was consideredstatistically significant.

RESULTS

Demographics of the subjects at baseline are shownin Table 1. The mean age of patients in the treatedgroup (n = 187) was 68 � 10.14 years as comparedwith 65.77 � 9.64 years in the untreated group(n = 100). There were no significant differencesbetween the two groups for age at baseline and dura-tion of follow up (independent samples t-test,P = 0.08, P = 0.25, respectively), whereas there weresignificant differences between the groups withrespect to gender ratio, and baseline values of IOPand CCT, as indicated in Table 1. The proportion offemales was significantly higher in the untreatedcontrol group when compared with the treatedgroup; however, a comparison of baseline CCTbetween males and females in both groups revealed

Antiglaucoma therapy and corneal thickness 349


no significant differences (P > 0.05). The treated eyeshad a significantly lower CCT (P = 0.01) and signifi-cantly higher IOP compared with the untreatedcontrol group (P < 0.0001), which was as expected.

An inverse but non-significant relationship wasfound between CCT and age in both treated(r = -0.052, P = 0.45) and untreated eyes (r = -0.171,P = 0.08). Linear regression analysis revealed areduction in CCT per decade by 4.5 mm and 5.5 mmin the treated and untreated eyes, respectively, andthis relationship between age and CCT has beenshown in Figure 1. A significant positive correlation(r = 0.32, P < 0.0001) was observed between baselineCCT and IOP (Fig. 2); however, no significant rela-tionship was found between CCT change and IOPchange (r = 0.07, P = 0.35). The distribution of diag-noses of patients according to various medications inthe treated group is shown in Table 2. POAG was thecommonest form of treated glaucoma, followed byNTG. No patient converted from NTG to POAGduring the course of the study.

A comparison of the CCT and IOP values at base-line and follow up for the two different types ofglaucoma among the treated eyes is shown inTable 3. There was a significant CCT reduction(P < 0.0001) in eyes with both NTG and POAG.Although the baseline CCT in eyes with NTG was

significantly lower than those with POAG (P = 0.01),there was no difference in the amount of CCT change(P = 0.61) between the two groups. The comparisonbetween CCT at baseline and CCT at follow up forboth control eyes and treated eyes classified intodifferent medication groups is shown in Table 4 andFigure 3.

There was no significant change in the control eyes(P > 0.05), whereas a significant reduction in CCT(P < 0.0001) was noted in the treated eyes, includingall medication groups except beta-blockers and the

Table 1. Baseline characteristics of treated and control eyes

Parameter Treated eyes (n = 187) Control eyes (n = 100) P-value

Age (years) 68 � 10.14 65.77 � 9.64 0.08Gender (male/female) 103/98 35/65 0.009Duration of follow up 6.92 � 1.67 years 6.58 � 1.93 years 0.25CCT (mm) 536.48 � 35.44 546.79 � 30.91 0.01IOP (mmHg) 21.68 � 4.64 17.33 � 3 <0.0001

CCT, central corneal thickness; IOP, intraocular pressure.

40–49 50–59 60–69 70–79 80–89450

500

550

600

CCT treated (P = 0.45) CCT controls (P = 0.08)

Age (years)

Ce

ntr

al co

rne

al th

ickn

ess

CC

T (

μm

)

Figure 1. Relationship between age and central corneal thick-ness. CCT, central corneal thickness.

400 450 500 550 600 650

0

10

20

30

40

CCT (μm)

IOP

(m

m H

g)

Figure 2. Relationship between intraocular pressure (IOP) andcentral corneal thickness (CCT).

Table 2. Distribution of diagnosis among treated eyes

Medication POAGn (%)

NTGn (%)

Totaln (%†)

PG 56 (69.1) 25 (30.9) 81 (43.3)BB 26 (76.5) 8 (23.5) 34 (18.2)PG + BB 36 (80) 9 (20) 45 (24.1)PG + BB + CAI 11 (100) 0 11 (5.9)PG + BB + AA 5 (83.3) 1 (16.7) 6 (3.2)Miscellaneous 9 (90) 1 (10) 10 (5.3)Total n (%‡) 143 (76.5) 44 (23.5) 187 (100)

†Column percent. ‡Row percent. AA, alpha agonists; BB, beta-blockers; CAI, carbonic anhydrase inhibitors; NTG, normaltension glaucoma; PG, prostaglandins; POAG, primary open-angle glaucoma.

350 Viswanathan et al.


small subgroup who were on a combination of pros-taglandins, beta-blockers and alpha agonists. Themean reduction in CCT was 12.29 � 13.65 mm inthe treated eyes and 1.17 � 8.75 mm in controls.The range of differences in CCT (present CCT minusbaseline CCT) was from -48 mm (thinned) to +35 mm(thickened). Of 187 eyes, 152 eyes (81.1%) becamethinner, 33 eyes (17. 9%) became thicker and two eyes(1%) remained exactly the same. Among treated eyes,56.1% eyes thinned more than 10 mm and 4.3% thick-ened more than 10 mm, whereas among controls, 14%eyes thinned more than 10 mm and 10% thickenedmore than 10 mm. There was no significant relation-ship noted between the amount of observed change in

CCT and the duration of follow up in years (r = -0.10,P = 0.16) in treated eyes.

DISCUSSION

CCT has emerged as an important parameter forglaucoma diagnosis and management. Our studywas designed to investigate whether CCT changesoccur over time in eyes with glaucoma treatedwith topical antiglaucoma medications. This studyhas shown that CCT decreased by a mean 12.29 �13.65 mm (2.26 � 2.52% reduction) in treated eyesover a follow-up period of 6.92 � 1.67 years. Amongthe treated eyes, the reduction in CCT was mostsignificant in those treated with either prostagland-ins or a combination of prostaglandins and beta-blockers, while there was no significant reduction ineyes treated only with beta-blockers (P = 0.16) whencompared with untreated control eyes. The meanreduction in CCT in the eyes treated with prostag-landins was 15.89 � 13.00 mm, amounting to a per-centage reduction rate of 2.91 � 2.37%. This isconsistent with the findings of other studies thathave shown that prostaglandins are associated witha significant reduction in CCT over time.13–18

Kim and Cho recently showed a statistically sig-nificant reduction in CCT by 5.4 mm after 24 monthsof treatment with latanoprost in patients with NTG.13

Sen et al. reported a percentage reduction rate in CCTof 1.9 � 2.4% and 2.8 � 1.8% during 24 months

Table 3. Comparison of CCT and IOP values

Eyesn (%)

BaselineCCT (mm)

CCT at followup (mm)

Difference(mm)

P-value Baseline IOP(mmHg)

IOP at followup (mmHg)

Difference(mmHg)

P-value

Treated187 (100%)

536.48 � 35.44 524.19 � 34.88 12.29 � 13.65 <0.0001 21.69 � 4.61 14.07 � 1.73 7.56 � 4.43 <0.0001

POAG143 (76.5%)

539.56 � 36.74 526.39 � 36.19 13.17 � 13.09 <0.0001 22.73 � 3.72 14.26 � 1.62 8.47 � 3.84 <0.0001

NTG44 (23.5%)

524.06 � 30.83 512.02 � 29.98 12.04 � 12.70 <0.0001 16.59 � 2.91 13.54 � 1.75 3.14 � 2.53 <0.0001

CCT, central corneal thickness; IOP, intraocular pressure; NTG, normal tension glaucoma; POAG, primary open-angle glaucoma.

Table 4. Comparison between central corneal thickness at baseline and at follow up

Eyes Baseline CCT(mm)

CCT at follow up(mm)

Difference(mm)

P-value

Control eyes (n = 100) 546.79 � 30.91 545.62 � 31.79 1.17 � 8.75 0.18Treated eyes (n = 187) 536.48 � 35.44 524.19 � 34.88 12.29 � 13.65 <0.0001PG (n = 82) 537.69 � 37.10 521.79 � 34.94 15.89 � 13.00 <0.0001BB (n = 40) 539.04 � 28.52 535.17 � 27.98 3.88 � 13.00 0.06PG + BB (n = 45) 530.10 � 35.50 515.97 � 34.97 14.13 � 13.19 <0.0001PG + BB + CAI (n = 9) 532.73 � 25.59 519.73 � 31.26 13.00 � 13.54 <0.0001PG + BB + AA (n = 4) 505.83 � 34.69 498.83 � 28.12 7.00 � 10.60 0.16Miscellaneous (n = 7) 568.20 � 33.48 558.60 � 34.60 9.60 � 9.55 0.004

AA, alpha agonists; BB, beta-blockers; CAI, carbonic anhydrase inhibitors; CCT, central corneal thickness; PG, prostaglandins.

Baseline Follow up 490

500

510

520

530

540

550

560Controls

PG

BB

PG & BB

PG & BB & CAI

PG & BB& AA

Me

an

CC

T (

μm)

Figure 3. Comparison between CCT at baseline and at followup between controls and different medication groups. AA, alphaagonists; BB, beta-blockers; CAI, carbonic anhydrase inhibitors;CCT, central corneal thickness; PG, prostaglandins.



of treatment with latanoprost and bimatoprost,respectively.17 A recent study also compared theeffects of latanoprost, travoprost and bimatoprost onCCT,14 and observed a statistically significant reduc-tion (P < 0.001) in CCT in all groups over17.19 � 15.71 months follow up, the reduction ofCCT being 14.95 � 5.04 mm, 15.73 � 3.25 mm and17.00 � 6.23 mm in the latanoprost, travoprost andbimatoprost groups, respectively. A reduction of CCTin patients under treatment with topical prosta-glandins when compared with topical carbonicanhydrase inhibitors has been reported,20 and Hara-symowycz et al. reported a 6.9-mm thinning of CCT atthe end of a 6-week follow up after treatment withtravoprost.16

Brandt et al.10 analyzed CCT change (mean 3.8years apart) in a large cohort of OHTS participants,and they observed that CCT decreased by a meanof 1.0 � 3.4 mm/year among participants originallyrandomized to observation, as compared with0.5 � 3.5 mm/year among all participants originallyrandomized to medication (P < 0.000), which isin contrast to our findings. However, when theyanalyzed participants treated only with topical pros-taglandin analogues, they found a greater rate ofdecrease in CCT per year than participants treatedonly with topical beta-blockers as found in thisstudy. We detected an overall CCT reductionrate of 2.1 � 2.4 mm/year among all treated eyes,2.8 � 2.5 mm/year among prostaglandin-treatedeyes, 0.9 � 2.1 mm/year among beta-blocker-treatedeyes and 0.2 � 1.6 mm/year among untreated con-trol eyes.

The strengths of the study by Brandt et al.10 werethe large sample size and the multivariate linearmodel to study CCT change that took into account allfive readings of the first and second CCT measure-ments; however, the limitation was that the observa-tion group also had exposure to topical medicationsfor a variable period of time (just over 1-year meanduration) between the two CCT measurements, andmost of these subjects were likely to have been onprostaglandins, whereas the original medicationgroup was predominantly on beta-blockers orothers. This in fact may explain why the OHTS datasuggested less thinning in the original treatmentgroup than the observation group – they had fewersubjects with prostaglandin therapy – rather thantreatment itself being protective to CCT. The prostag-landin changes could easily have occurred within 1year of therapy. In addition, the OHTS data includedonly ocular hypertensive subjects with a mean CCTof around 573 mm, many of whom never developedclinical glaucoma.

The strengths of our study include a large sample,long follow up of nearly 7 years, minimum of 3 yearson drug, analysis of treated eyes belonging to both

POAG and NTG glaucoma subtypes, and an age-matched control group with no prior exposure totopical medications. As medications were only usedfrom baseline, the pretreatment CCT values werereliable and free from former antiglaucoma medica-tion effects. We found no time-related effect on theamount of CCT change; this is similar to Zhong et al.,who reported no significant difference in the amountof reduction in CCT between the patients with lessthan or equal to 6 months treatment and the patientswith more than 6 months of treatment in the threeprostaglandin groups.14 This suggests that the CCTthinning caused by prostaglandins mainly occurswithin the initial treatment period. Our findingswould support this as cross-sectional analysis ofduration of treatment versus the amount of thinningdid not reveal any relationship, suggesting that thechange is not progressive at least after 3 years oftreatment.

Interestingly, there was no significant CCT reduc-tion in the group of eyes treated with a combinationof prostaglandins, beta-blockers and alpha agonists.Grueb et al.21 recently described a reversible increasein CCT after topical administration of brimonidine0.1%, which could mean that the CCT reductioncaused by prostaglandins might have been compen-sated by an increase in CCT caused by the alphaagonists, effectively resulting in no significantchange in CCT in that group.

A potential mechanism for prostaglandins toinduce thinning could be their effect on matrixmetalloproteinases (MMPs). MMPs are a group ofenzymes that degrade collagen, a key constituent ofbasement membrane and extracellular matrix com-ponents.22 The MMP family includes about 20 typesof enzymes, which may be found in the anteriorsegment of the eye, including the corneal epithelium,stroma and endothelium, conjunctiva, lacrimal film,aqueous humour, trabecular meshwork, ciliarymuscle cells, and lens.23–26 The IOP lowering effect ofprostaglandin analogues is known to be caused byMMP activation in the smooth muscle of the ciliarybody.27The overactivity of MMP-2 in the corneainduced by prostaglandins may progressively thinthe cornea in pathological conditions, such askeratoconus.28

Prostaglandins appear to promote collagen gelcontraction and affect the corneal stromal structureby decreasing the extracellular matrices, and thusmight affect the CCT. Liu et al. compared the effects ofthe antiglaucoma drugs latanoprost and timolol onthe contraction of corneal fibroblasts cultured ina three-dimensional collagen gel; they reportedthat latanoprost stimulated collagen gel contractionmediated by corneal fibroblasts in a concentration-and time-dependent manner, whereas timolol hadno such effect.29 They postulated that this action



might affect corneal shape and thickness, andthereby influence measurement of IOP. The cornealstroma is mainly composed of collagen fibres, andthe reduction in CCT caused by prostaglandinsappears to result from their upregulation of localMMPs.

Earlier cross-sectional studies6–8 demonstratedthat thinner corneas are associated with increasingage. Aghaian et al7 showed a 3-mm decrease in CCTper decade of age in a multiracial patient cohort,while the OHTS demonstrated a 6.3-mm decrease inCCT per decade of age.5 In our study, linear regres-sion analysis revealed a reduction in CCT per decadeof 4.5 mm and 5.5 mm in the treated and untreatedeyes, respectively; this agrees with these studies.

In terms of clinical implications, the change inCCT is not likely to be of much clinical significance,although this study was not designed specifically toassess this aspect. The OHTS data also suggested thatthe changes they noted over time were unlikely toinfluence clinical decisions. Ehlers et al.3 had demon-strated a 5-mmHg change in IOP for 70 mm of CCTvariation, while Doughty and Zaman,1 in their meta-analysis, had reported a change of IOP by 2.5 mmHgper 50 mm of CCT variation. In our study, prostag-landin analogues decreased CCT values by approxi-mately 16 mm over a mean 6.9 years, resulting infalsely lower IOP values by 1.14 and 0.8 mmHgaccording to Ehlers and Doughty’s IOP correctionformulas, respectively; these are not likely to beclinically relevant. However, some individualsshowed much greater changes of up to 48 mm ofthinning, and this could impact Goldmann applana-tion tonometry measurement accuracy.

The limitations of our study included the fact thatthe ultrasonic pachymeter was upgraded in 2005,and so some patients underwent CCT measure-ment using two different ultrasonic pachymeters.However, we found no significant difference(P = 0.19) in the amount of CCT change between twogroups with baseline CCT measured before or after2005. In addition, although the baseline pachymetrymeasurements were performed by different opera-tors, CCTs at the final visits were measured by thesame operator, and earlier studies have indicatedgood intra- and interobserver reproducibility forultrasonic CCT measurements.30,31

Our study had insufficient numbers to examinethe effects of the three different prostaglandins indi-vidually; the majority of patients were on latano-prost. The study may also be limited by the exclusionof subjects undergoing laser and surgery. Thesegroups typically would have been showing eitherhigher IOPs or progressive change, and potentiallymay have included a larger proportion of patientswith thinner corneas or with greater thinning overtime. However, we felt there was the potential for

both laser and surgery of any kind to impact on thecorneal endothelium and its function, and thereforechose not to include them. Furthermore, althoughthe ‘control’ group of suspects who are not on treat-ment does not represent true normal controls, it doesrepresent an untreated group, none of whom con-verted to glaucoma during follow up. Ideally, weshould have followed 100 normal subjects with norisk factors, but we did not have such a sample withbaseline data available.

In conclusion, long-term treatment with prostag-landin analogues is associated with a small but sig-nificant thinning of the central cornea. The potentialimplication on IOP measurements should be consid-ered in glaucoma practice, but in most cases is notlikely to be clinically relevant. Given the knownthinning with age, repeat CCT measurements areappropriate in long-term glaucoma management.

REFERENCES

1. Doughty MJ, Zaman ML. Human corneal thicknessand its impact on intraocular pressure measures: areview and meta-analysis approach. Surv Ophthalmol2000; 44: 367–408.

2. Whitacre MM, Stein RA, Hassanein K. The effect ofcorneal thickness on applanation tonometry. Am J Oph-thalmol 1993; 11: 592–6.

3. Ehlers N, Bramsen T, Sperling S. Applanation tonom-etry and central corneal thickness. Acta Ophthalmol(Copenh) 1975; 53: 34–43.

4. Stodtmeister R. Applanation tonometry and correctionaccording to corneal thickness. Acta Ophthalmol Scand1998; 76: 319–24.

5. Brandt JD, Beiser JA, Kass MA et al. Ocular Hyperten-sion Treatment Study (OHTS) Group. Central cornealthickness in the Ocular Hypertension Treatment Study(OHTS). Ophthalmology 2001; 108: 1779–88.

6. Nemesure B, Wu SY, Hennis A et al. Corneal thicknessand intraocular pressure in the Barbados eye studies.Arch Ophthalmol 2003; 121: 240–4.

7. Aghaian E, Choe JE, Lin S et al. Central corneal thick-ness of Caucasians, Chinese, Hispanics, Filipinos,African Americans, and Japanese in a glaucoma clinic.Ophthalmology 2004; 111: 2211–19.

8. Foster PJ, Baasanhu J, Alsbirk PH et al. Central cornealthickness and intraocular pressure in a Mongolianpopulation. Ophthalmology 1998; 105: 969–73.

9. Weizer JS, Stinnett SS, Herndon LW. Longitudinalchanges in central corneal thickness and their relationto glaucoma status: an 8 year follow up study. Br JOphthalmol 2006; 90: 732–6.

10. Brandt JD, Gordon MO, Beiser JA et al. Ocular Hyper-tension Treatment Study Group. Changes in centralcorneal thickness over time. Ophthalmology 2008; 115:1550–6.

11. Inoue K, Okugawa K, Kato S et al. Ocular factors rel-evant to anti-glaucomatous eyedrop-related keratoepi-theliopathy. J Glaucoma 2003; 12: 480–5.



12. Kaminski S, Hommer A, Koyuncu D et al. Influence ofdorzolamide on corneal thickness, endothelial cellcount and corneal sensibility. Acta Ophthalmol Scand1998; 76: 78–9.

13. Kim HJ, Cho BJ. Long-term effect of latanoprost oncentral corneal thickness in normal tension glaucoma.J Ocul Pharmacol Ther 2011; 27: 73–6.

14. Zhong Y, Shen X, Yu J et al. The comparison of theeffects of latanoprost, travoprost, and bimatoprost oncentral corneal thickness. Cornea 2011; 30: 861–4.

15. Stefan C, Dumitrica DM, Tebeanu E et al. Prostaglan-din analogues and central corneal thickness. Oftalmo-logica 2007; 51: 95–9.

16. Harasymowycz PJ, Papamatheakis DG, Ennis M et al.Relationship between travoprost and central cornealthickness in ocular hypertension and open-angle glau-coma. Cornea 2007; 26: 34–41.

17. Sen E, Nalcacioglu P, Yazici A et al. Comparison of theeffects of latanoprost and bimatoprost on centralcorneal thickness. J Glaucoma 2008; 17: 398–402.

18. Hatanaka M, Vessani RM, Elias IR et al. The effect ofprostaglandin analogs and prostamide on centralcorneal thickness. J Ocul Pharmacol Ther 2009; 25: 51–3.

19. Bafa M, Georgopoulos G, Mihas C et al. The effect ofprostaglandin analogues on central corneal thicknessof patients with chronic open-angle glaucoma: a 2-yearstudy on 129 eyes. Acta Ophthalmol 2011; 89: 448–51.

20. Viestenz A, Martus P, Schlötzer-Schrehardt U et al.Impact of prostaglandin-F (2alpha)-analogues and car-bonic anhydrase inhibitors on central corneal thick-ness – a cross-sectional study on 403 eyes. Klin MonatsblAugenheilkd 2004; 221: 753–6.

21. Grueb M, Mielke J, Rohrbach JM et al. Effect of bri-monidine on corneal thickness. J Ocul Pharmacol Ther2011; 27: 503–9.

22. Leonardi A, Brun P, Abatangelo G et al. Tear levels andactivity of matrix metalloproteinase (MMP)-1 andMMP-9 in vernal keratoconjunctivitis. Invest Ophthal-mol Vis Sci 2003; 44: 3052–8.

23. Gaton DD, Sagara T, Lindsey JD et al. Increased matrixmetalloproteinase 1, 2 and 3 in the monkey uveoscleraloutflow pathway following topical prostaglandin F2a-isopropyl ester treatment. Arch Ophthalmol 2001;119: 1165–70.

24. Zhong Y, Gao J, Ye W et al. Effect of latanoprost acidand pilocarpine on cultured rabbit ciliary muscle cells.Ophthalmic Res 2007; 39: 232–40.

25. Ocklind A. Effect of latanoprost on the extracellularmatrix of the ciliary muscle: a study on cultured cellsand tissue sections. Exp Eye Res 1998; 67: 179–91.

26. Weinreb RN, Kashiwagi K, Kashiwagi F et al. Prostag-landins increase matrix metalloproteinase release fromhuman ciliary smooth muscle cells. Invest Ophthalmol VisSci 1997; 38: 2772–80.

27. Ooi YH, Oh D-J, Rhee DJ et al. Effect of bimatoprost,latanoprost, and unoprostone on matrix metalloprotei-nases and their inhibitors in human ciliary bodysmooth muscle cells. Invest Ophthalmol Vis Sci 2009; 50:5259–65.

28. Smith VA, Hoh HB, Littleton M et al. Over-expressionof gelatinase A activity in keratoconus. Eye 1995; 9:429–33.

29. Liu Y, Yanai R, Lu Y et al. Effects of antiglaucoma drugson collagen gel contraction mediated by humancorneal fibroblasts. J Glaucoma 2006; 15: 255–9.

30. Miglior S, Albe E, Guareschi M et al. Intraobserverand interobserver reproducibility in the evaluationof ultrasonic pachymetry measurements of centralcorneal thickness. Br J Ophthalmol 2004; 88: 174–7.

31. Marsich MW, Bullimore MA. The repeatability ofcorneal thickness measures. Cornea 2000; 19: 792–5.



longitudinal effect of topical antiglaucoma medications on central corneal thickness

Documents