Long-Term Effect of Latanoprost on Central CornealThickness in Normal Tension Glaucoma

Hyung Jun Kim and Byung Joo Cho

Abstract

Purpose: To evaluate the long-term effect of latanoprost on central corneal thickness (CCT) in patients withnormal tension glaucoma (NTG).Methods: This was a retrospective study and included 166 eyes of 166 patients [128 with NTG and 38 withglaucoma suspect, suspicious discs with normal visual fields, and an intraocular pressure (IOP) �21 mmHg asthe control group]. Patients with newly diagnosed NTG and who had not had previous topical glaucomatoustreatment were followed �24 months and received latanoprost 0.005% monotherapy once a day. CCT mea-surements were performed with an ultrasound pachymeter. CCT measurements before treatment and 24 monthsafter treatment were analyzed.Results: There were no significant differences between the latanoprost group and the control group with respectto sex, age, baseline IOP, and CCT. A statistically significant reduction in the mean CCT was observed in thelatanoprost group [535.5� 37.9 vs. 530.1� 36.4 mm (n¼ 128), P< 0.01], but not in the control group [543.1� 40.2vs. 542.6� 37.0 mm (n¼ 38), P¼ 0.786].Conclusions: Long-term use of latanoprost may decrease the CCT in patients with NTG. Therefore, cliniciansmust be aware of longitudinal CCT variations that may arise throughout the follow-up period for proper IOPtargeting and management.

Introduction

Glaucoma is the second leading cause of irreversiblevision loss worldwide, and is responsible for *15% of

cases of blindness.1 Intraocular pressure (IOP) is one of themost important parameters in the diagnosis and treatment ofglaucoma. The gold standard for IOP measurement is theGoldmann applanation tonometer; however, this device maybe influenced by central corneal thickness (CCT).2 CCT hasbecome an important factor in the management of glaucomaafter understanding its influence on IOP measurements. Theimportance of CCT as a potential risk factor for glaucomaand ocular hypertension has also been studied. The OcularHypertension Treatment Study, which focused on risk fac-tors for glaucomatous visual field (VF) loss and optic nervehead changes in patients with ocular hypertension, sug-gested that the CCT is the strongest predictive factor.3 Morerecent studies have suggested that CCT may be an inde-pendent risk factor for progression in patients with open-angle glaucoma as well.4

Normal tension glaucoma (NTG) represents a subgroup ofpatients with open-angle glaucoma with measured, un-

treated IOP in the statistically normal range. Typically, theprevalence of NTG is much higher in Korea and Japan thanin Western countries. It is widely believed that IOP plays arole in the pathogenesis of NTG.5–8 In patients with NTG, theCollaborative Normal-Tension Glaucoma Trial demonstratedthat a 30% reduction in IOP after treatment could help pre-vent further VF loss in patients in the United States.9

Latanoprost was introduced in the United States in 1996 asthe first prostaglandin-based IOP-lowering medication.Prostaglandin analogs reduce IOP by enhancing the aqueoushumor through the uveoscleral pathway to the suprachor-oidal space and episcleral veins. Although the exact mecha-nism of action of prostaglandin analogs is still not fullyunderstood, it has been theorized that they act primarilythrough the activation of matrix metalloproteinases (MMPs)10

and a reduction of different collagen types in the eye.11 Be-cause the cornea is mainly composed of collagen fibers andtaking into account the importance of corneal thickness in theevaluation of glaucoma, it is important to determine the CCTvalues serially during the follow-up of glaucoma patientsbecause of the longitudinal changes that might occur owing tothe medication.

This study was presented as a poster at the ARVO meeting, May 2–6, 2010, Fort Lauderdale, Florida.Department of Ophthalmology, Konkuk University Medical Center, Konkuk University School of Medicine, Seoul, Republic of Korea.

JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICSVolume 27, Number 1, 2011ª Mary Ann Liebert, Inc.DOI: 10.1089/jop.2010.0071

73

Prostaglandin analogs have been shown to decrease theCCT values in some studies,12–14 but previous studies ofprostaglandin analogs had mostly short-term follow-up pe-riod and included patients with various type of glaucoma.There was no study including only NTG patients with long-term follow-up period. Thus, this study was performed toevaluate the long-term effect of latanoprost on the CCT inpatients with NTG.

Methods

This was a retrospective study and included 166 eyes of166 patients (128 with NTG and 38 with glaucoma suspect).The study was approved by the Institutional Review Boardof Konkuk University Medical Center and conformed to theDeclaration of Helsinki. The patients’ records were reviewedand consecutive patients meeting the inclusion/exclusioncriteria were enrolled in the study. The inclusion criteriawere patients with newly diagnosed NTG who had not re-ceived previous topical glaucomatous treatment and werefollowed �24 months during latanoprost 0.005% mono-therapy once a day. NTG was defined as open angle basedon gonioscopy as follows: an IOP �21 mmHg, and a typicalglaucomatous optic nerve defect and/or glaucomatous VFdefect. The inclusion criteria for the control group were pa-tients with glaucoma suspect, suspicious discs with normalVFs, and an IOP �21 mmHg. The patients with corneal dis-eases, and histories of trauma, intraocular surgery, an ocularinfection, contact lens use, and corticosteroid use were ex-cluded from the study.

CCT measurements were performed using an ultrasoundpachymeter (AL-2000; Bio & Pachymeter�) by the samephysician recording a mean of 5 consecutive readings. Oneeye of each patient was randomly selected for statisticalanalysis and the mean of 5 consecutive measurements wasconsidered for analysis. CCT measurements before treatmentand 24 months after treatment were analyzed.

To evaluate the variation in CCT values at different timeintervals within the groups, a paired-sample t-test was used.An independent sample t-test was used to compare the meanage, baseline IOP, and baseline CCT between groups. A Fi-scher w2 test was performed to evaluate sex difference be-

tween the 2 groups. A P value <0.05 was consideredstatistically significant. All the statistical analyses were car-ried out using SPSS 12.0.

Results

The mean (�standard deviation) age of the patients was55.9� 13.6 years in the latanoprost group (n¼ 128) and55.6� 12.8 years in the control group (n¼ 38). There were nosignificant differences between the groups with respect tosex, age, and baseline values of IOP and CCT, as indicated inTable 1. The CCT values and changes in both groups frombaseline to the 24th month are shown in Table 2. The CCT ofthe latanoprost group after 24 months of treatment wassignificantly thinner than the baseline value (535.5� 37.9 vs.530.1� 36.4 mm, P< 0.01), but there was no statistically sig-nificant change in the CCT in the control group (543.1� 40.2vs. 542.6� 37.0 mm, P¼ 0.786).

Discussion

The purpose of this study was to determine the long-termeffect of latanoprost on the CCT in patients with NTG. NTGaccounts for a significant percentage of all forms of open-angle glaucoma. In Western countries, the incidence of NTGis believed to be generally much lower, and occurs in anolder population than that of patients with primary open-angle glaucoma (POAG).15 In contrast, Shiose et al.16 havereported NTG in 2.04% and POAG in 0.58% of the adultJapanese population. Also, a study by Iwase et al.17 involv-ing 3,021 persons >40 years in Tajimi City, Japan, revealed aprevalence of POAG of 3.9% in which 92% had an IOP�21 mmHg. A similar distribution has been noted in Korea.18

Japan and Korea may have a similar incidence of NTG be-cause of the close genetic ties between the countries.19–21

Thus, we determined the long-term effect of latanoproston CCT values with a longitudinal follow-up of 24 months inpatients with NTG. This study showed a 5.4 mm statisticallysignificant reduction in CCT after 24 months of treatmentwith latanoprost (Fig. 1). The result is consistent with theobservations noted for other prostaglandin analogs in otherstudies.22–25

Table 1. Baseline Characteristics

Latanoprost group (n¼ 128) Control group (n¼ 38) P value

Sex (M/F) 59/69 14/24 0.671Age (years, mean� SD) 55.9� 13.6 55.6� 12.8 0.978Baseline IOP (mmHg, mean� SD) 16.0� 2.9 16.1� 2.9 0.893Baseline CCT (mm, mean� SD) 535.5� 37.9 543.1� 40.2 0.288

CCT, central corneal thickness; IOP, intraocular pressure; SD, standard deviation.

Table 2. Mean� Standard Deviation and Changes of Central Corneal Thickness Values at the Baseline

and Follow-Up for Both Groups

CCT (mm) Baseline (mean� SD) 24 months (mean� SD) CCT change (mm) P value

Latanoprost group (n¼ 128) 535.5� 37.9 530.1� 36.4 �5.40� 10.92 <0.01a

Control group (n¼ 38) 543.1� 40.2 542.6� 37.0 �0.53� 11.72 0.786

Statistical significance was evaluated by paired sample t-test.aStatistically significant.

74 KIM AND CHO

Stefan et al.22 described equivalent CCT thinning in patientsreceiving travoprost or latanoprost over a period of 3 months.Viestenz et al.13 also demonstrated a reduction in the CCT inpatients under treatment with topical prostaglandin F2-alphacompared with topical carbonic anhydrase inhibitors. Har-asymowycz et al.23 studied the effect of travoprost on the CCTand reported a 6.9mm thinning of the mean CCT at the end ofa 6-week follow-up. Sen et al.24 reported a reduction in theCCT of 1.9%� 2.4% and 2.8%� 1.8% during 24 months oftreatment with latanoprost and bimatoprost, respectively.Hatanaka et al.25 showed that topical therapy with prosta-glandin analogs (latanoprost and travoprost) and bimatoprostwas associated with a reduction in the CCT over a period of atleast 8 weeks. Previous studies have shown some differencesin CCT reduction because of the differences in follow-up pe-riods and in the patients involved in the study. This studyincluded only NTG patients and showed more thinner base-line CCT than past studies included mainly POAG patients,corresponding some studies showed that NTG patients havelower CCT measurements.26–28 We believe that differentbaseline CCT might affect CCT reduction. The other possibleexplanations for differences in CCT reduction are that CCTmeasurements might be affected by some factors such as theinstruments used, interobserver variations, age, sex, race, anddiurnal variation. Despite this difference, we confirmed thatprostaglandin analogs could reduce the CCT in every study(Table 3).

MMPs are proteases that degrade collagen, a basementmembrane constituent, and other components in the extra-cellular matrix.29 MMPs are important in tissue remodelingand wound healing.30 The MMPs family includes about 20types of enzymes that may be found in the anterior segmentof the eye, including lacrimal film, corneal epithelium, stro-ma and endothelium, aqueous humor, lens, trabecularmeshwork, and conjunctiva. It is known that MMPs arefound in the cornea, and overactivity of MMP-2 may be re-

lated to pathologic conditions such as keratoconus.31 Thepredominant IOP effect of prostaglandin analogs is known tobe caused by MMP-1 activation in the smooth muscle of theciliary body,32 whereas the corneal effects may be mediatedby MMP-2 crossactivation. Thus, it is possible that throughdirect activation of the corneal MMPs, prostaglandin analogsinduce physical changes to the corneal architecture.

Ehlers et al.33 have found 5 mmHg change in IOP for 70mmof CCT variation. Doughty and Zaman,34 in their meta-analysis, have reported the relationship as 2.5 mmHg IOPchange per 50mm CCT variation. In our study, the smallamount of corneal thinning, although statistically significant,may not be clinically significant and may not influence IOPmeasurements. However, because long-term use of prosta-glandin analogs may increase CCT reduction, we beleive thatthe increase of corneal thinning may influence increasinglyIOP measurements. Thus, while setting the target IOP levels inglaucoma practice, the falsely lower IOP values owing to thethinning effects of prostaglandin analogs have to be taken intoaccount. CCT evaluation along with IOP measurements at allfollow-up visits may be beneficial for this purpose. The limi-tation of this study was the retrospective design and the rel-atively small control group. To determine the magnitude ofthe effect on the CCT change by prostaglandin analogs on IOPin the long-term, further prospective, large-scale, multi-centered studies with longer follow-up comparing all prosta-glandin analogs with a control group are needed.

Conclusions

Long-term use of latanoprost may decrease the CCT inpatients with NTG. Therefore, clinicians must be aware ofthe longitudinal CCT variations that may arise through-out the follow-up period for proper IOP targeting andmanagement.

Acknowledgment

This work was supported by the Konkuk UniversityMedical Center Research Grant 2011.

Author Disclosure Statement

No competing financial interests exist.

References

1. Quigley, H.A. Number of people with glaucoma worldwide.Br. J. Ophthalmol. 80:389–393, 1996.

FIG. 1. Longitudinal changes in central corneal thickness.

Table 3. Previous Studies of Other Prostaglandin Analogs

Study Design Sample size Patients Follow-up period Treatments Mean CCT reduction (mm)

Hatanaka et al.25 Prospective 73 POAG 8 weeks Latanoprost �4.69Travoprost �6.22Bimatoprost �4.06

Sen et al.24 Prospective 94 POAG, PSXG,NTG, OHT

24 months Latanoprost �6.7Bimatoprost �7.7

Stefan et al.22 Prospective 26 POAG, OHT 3 months Latanoprost �4.2Travoprost �6.23

Harasymowycz et al.23 Prospective 379 POAG, OHT 6 weeks Travoprost POAG: �7.59OHT: �5.81

NTG, normal tension glaucoma; OHT, ocular hypertension; POAG, primary open angle glaucoma; PSXG, pseudoexfoliation glaucoma.

LONG-TERM EFFECT OF LATANOPROST ON CCT IN NTG 75

2. Goldmann, H., and Schmidt, T. About applanation tonom-etry. Ophthalmologica 134:221–242, 1957.

3. Gordon, M.O., Beiser, J.A., Brandt, J.D., et al. The ocularhypertension treatment study: baseline factors that predictthe onset of primary openangle glaucoma. Arch. Ophthalmol.120:714–720, 2002.

4. Herndon, L.W., Weizer, J.S., and Stinnett, S.S. Central cor-neal thickness as a risk factor for advanced glaucomadamage. Arch. Ophthalmol. 122:17–21, 2004.

5. Cartwright, M.J., and Anderson, D.R. Correlation of asym-metric damage with asymmetric intraocular pressure innormal tension glaucoma (low-tension glaucoma). Arch.Ophthalmol. 106:888–900, 1988.

6. Crichton, A., Drance, S.M., Douglas, G.R., et al. Unequalintraocular pressure and its relation to asymmetric visualfield defects in low-tension glaucoma. Ophthalmology96:1312–1314, 1989.

7. Haefliger, I.O., and Hitchings, R.A. Relationship betweenasymmetry in visual field defects and intraocular pressuredifference in an untreated normal (low) tension glaucomapopulation. Acta Ophthalmol. 68:564–567, 1990.

8. Jonas, J.B., Grundler, A.E., and Gonzales-Cortes, J. Pressure-dependent neuroretinal rim loss in normal-pressure glau-coma. Am. J. Ophthalmol. 125:137–144, 1998.

9. Collaborative Normal-Tension Glaucoma Study Group. Theeffectiveness of intraocular pressure reduction in the treat-ment of normal-tension glaucoma. Am. J. Ophthalmol.126:498–505, 1998.

10. Schachtschabel, U., Lindsey, J.D., and Weinreb, R.N. Themechanism of action of prostaglandins on uveoscleral out-flow. Curr. Opin. Ophthalmol. 11:112–115, 2000.

11. Sagara, T., Gaton, D.D., Lindsey, J.D., et al. Topical prosta-glandin F2a treatment reduces collagen types I, III, and IV inthe monkey uveoscleral outflow pathway. Arch. Ophthalmol.117:794–801, 1999.

12. Arcieri, E.S., Pierre Filho, P.T., Wakamatsu, T.H., et al. Theeffects of prostaglandin analogues on the blood aqueousbarrier and corneal thickness of phakic patients with pri-mary open-angle glaucoma and ocular hypertension. Eye22:179–183, 2008.

13. Viestenz, A., Martus, P., Schlotzer-Schrehardt, U., et al. Im-pact of prostaglandin-F (2alpha)-analogues and carbonicanhydrase inhibitors on central corneal thickness—a cross-sectional study on 403 eyes. Klin. Monatsbl. Augenheilkd.221:753–756, 2004.

14. Lass, J.H., Eriksson, G.L., Osterling, L., et al.; LatanoprostCorneal Effects Study Group. Comparison of the cornealeffects of latanoprost, fixed combination latanoprost-timolol,and timolol: a double-masked, randomized, one-year study.Ophthalmology 108:264–271, 2001.

15. Stewart, W.C. Clinical Practice of Glaucoma. Thorofare, NJ:SLACK, Inc.; 1990; pp. 141–144.

16. Shiose, Y., Kitazawa, Y., Tsukahara, S., et al. Epidemiologyof glaucoma in Japan—a nationwide glaucoma survey. Jpn.J. Ophthalmol. 35:133–155, 1991.

17. Iwase, A., Suzuke, Y., Araie, M., et al. The prevalence ofprimary open-angle glaucoma in Japanese: the Tajimi Study.Ophthalmology 111:1641–1648, 2004.

18. Choe, Y.J. The prevalence of glaucoma in Korean adults. J.Korean Ophthalmol. Soc. 34:65–69, 1993.

19. Omoto, K., and Saitou, N. Genetic origins of the Japanese: apartial support for the dual-structure hypothesis. Am. J.Phys. Anthropol. 102:437–446, 1997.

20. Tang, S., Toda, Y., Kashiwagi, K., et al. The association be-tween Japanese primary open-angle glaucoma and normal

tension glaucoma patients and the optineurin gene. Hum.Genet. 113:276–279, 2003.

21. Woo, S.J., Kim, D.M., Kim, J.Y., et al. Investigation of theassociation between OPA1 polymorphisms and normal-tension glaucoma in Korea. J. Glaucoma. 13:492–495, 2004.

22. Stefan, C., Dumitrica, D.M., Tebeanu, E., et al. Prostaglandinanalogues and central corneal thickness. Oftalmologia 51:95–99, 2007.

23. Harasymowycz, P.J., Papamatheakis, D.G., Ennis, M., et al.For the Travoprost Central Corneal Thickness Study Group.Relationship between travoprost and central corneal thick-ness in ocular hypertension and open-angle glaucoma. Cor-nea 26:34–41, 2007.

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

25. Hatanaka, M., Vessani, R.M., Elias, I.R., et al. The effect ofprostaglandin analogs and prostamide on central cornealthickness. J. Ocul. Pharmacol. Ther. 25:51–53, 2009.

26. Morad, Y., Sharon, E., Hefetz, L., et al. Corneal thickness andcurvature in normal-tension glaucoma. Am. J. Ophthalmol.126:326–328, 1998.

27. Emara, B.Y., Tingey, D.P., Probst, L.E., et al. Central cornealthickness in low-tension glaucoma. Can. J. Ophthalmol.34:319–324, 1999.

28. Rufer, F., Westphal, S., and Erb, C. Comparison of centraland peripheral corneal thicknesses between normal subjectsand patients with primary open angle glaucoma, normaltension glaucoma and pseudoexfoliation glaucoma. Klin.Monatsbl. Augenheilkd. 224:636–640, 2007.

29. Leonardi, A., Brun, P., Abatangelo, G., et al. Tear levels andactivity of matrix metalloproteinase (MMP)-1 and MMP-9 invernal keratoconjunctivitis. Invest. Ophthalmol. Vis. Sci.44:3052–3058, 2003.

30. Hanemaaijer, R., Visser, H., Konttinen, Y.T., et al. A noveland simple immunocapture assay for determination ofgelatinase-B (MMP-9) activities in biological fluids: saliva frompatients with Sjogren’s syndrome contain increased latent andactive gelatinase-B levels. Matrix Biol. 17:657–665, 1998.

31. Smith, V.A., Hoh, H.B., Littleton, M., et al. Overexpression ofgelatinase A activity in keratoconus. Eye 9:429–433, 1995.

32. Wang, N., Lindsey, J.D., Angert, M., et al. Latanoprost andmatrix metalloproteinase-1 in human choroid organ cul-tures. Curr. Eye Res. 22:198–207, 2001.

33. Ehlers, N., Bramsen, T., and Sperling, S. Applanation to-nometry and central corneal thickness. Acta Ophthalmol.53:34–43, 1975.

34. Doughty, M.J., and Zaman, M.L. Human corneal thicknessand its impact on intraocular pressure measures: a reviewand meta-analysis approach. Surv. Ophthalmol. 44:367–408,2000.

Received: June 3, 2010

Accepted: October 11, 2010

Address correspondence to:Dr. Byung Joo Cho

Department of OphthalmologyKonkuk University Medical Center

Konkuk University School of Medicine#4-12 Hwayang-dong, Gwangjin-gu

Seoul 143-729Republic of Korea

E-mail: bjcho@kuh.ac.kr

76 KIM AND CHO