Received: 16 December 1999Revised: 20 March 2000Accepted: 21 March 2000

Abstract Purpose: To examine thecentral corneal thickness (CCT) andcorneal endothelium in both eyes ofpatients with unilateral exfoliationsyndrome (EXS). To determine theeffect of CCT on the measurement of intraocular pressure (IOP). Methods: In this cross-sectional clin-ical study, comparisons were madeof CCT (Humphrey UltrasonicPachometer), corneal endothelialcells (Keeler-Konan contact specularmicroscope) and IOP (Goldmann applanation tonometer) between theexfoliative (E) and fellow non-exfoliative (NE) eyes in 40 normo-tensive patients with unilateral EXS.The CCT was used to obtain a cor-rected value for the IOP. Results: TheE eyes had significantly higher val-ues for CCT (0.528±0.030 vs0.523±0.032 mm, P<0.01) and IOP(15.7±3.6 vs 14.4±2.9 mmHg,

P<0.001) than the fellow NE eyes.The paired E and NE eyes did notdiffer in endothelial cell density(2779±540 vs 2870±386 cells/mm2),in the coefficient of variation of cellsize (0.25±0.03 vs 0.26±0.03) or inthe frequency of hexagonal cells(80.5±6.5 vs 82.0±5.0%). After correcting IOP for CCT, the E eyesstill had significantly higher IOPthan the NE eyes (15.1±4.4 vs 14.2±3.7 mmHg, P<0.05). Conclusion: Normotensive eyes withearly EXS did not have quantitative(cell density) or qualitative (varia-tion in cell size, frequency of hexag-onal cells) morphological changes incorneal endothelium, but had highervalues for IOP and CCT. After cor-recting IOP for CCT, the E eyes stillhad significantly higher IOP than thefellow NE eyes.

Graefe’s Arch Clin Exp Ophthalmol (2000)238:659–663 © Springer-Verlag 2000 C L I N I C A L I N V E S T I G AT I O N

Päivi PuskaKristiina VasaraMika HarjuKirsi Setälä

Corneal thickness and corneal endothelium in normotensive subjects with unilateral exfoliation syndrome

Introduction

The clinical corneal manifestations of the exfoliationsyndrome (EXS) include deposition of pigment dis-persed from the iris pigment epithelium and of exfolia-tion material, on the posterior corneal surface [21].

Abnormalities of the iris vessels cause tissue hypoxiathat results in neovascularization and leakage of the vessels [17, 33]. Increased permeability of the blood–aqueous barrier with changes in the composition of theaqueous [1, 13, 16] is thought to affect the metabolism ofthe corneal endothelium.

In earlier specular and electron microscopic studies,reduced endothelial cell densities and morphological

changes in cell size (polymegethism) and cell shape(pleomorphism) have been demonstrated in exfoliative(E) eyes with or without glaucoma, and also in fellownon-exfoliative (NE) eyes [20, 29, 34, 36, 37]. Thechanges in the endothelium are supposed to be an earlyand essential sign of the syndrome [20, 29]. A real exfo-liation keratopathy has been described in advanced EXS,and the corneal endothelium has even been suggested tosynthesize the exfoliation material. [23, 24, 29].

The aim of the present study was to examine the cen-tral corneal thickness (CCT) and the corneal endotheli-um in non-glaucomatous, normotensive patients withearly unilateral EXS. The effect of CCT on intraocularpressure (IOP) readings was also determined.

P. Puska (✉ ) · K. Vasara · M. Harju K. SetäläHelsinki University Eye Hospital, Haartmanink.4, PL 220, 00029 HYKS,Helsinki, Finlande-mail: paivi.puska@huch.fiFax: +358-9-4715100

Subjects and methods

The study comprised 40 non-glaucomatous, normotensive (IOP≤22 mmHg) subjects (age 68.4±7.1 years, range 52–80 years) withunilateral EXS. Eyes previously subjected to ocular surgery or oc-ular trauma or with corneal opacities or signs of ocular inflamma-tion were not included. The subjects did not wear contact lenses,use ocular medication or have systemic or ocular diseases associ-ated with changes in the structure of the corneal endothelium.

Unilateral EXS was defined as the presence of biomicroscopi-cally detectable exfoliation material on the anterior lens capsule orat the pupillary border only in the E eye after pupillary dilatationwith 10% phenylephrine hydrochloride. Exfoliation material waspresent in 20 right and 20 left eyes. The signs related to pigmentdispersion were recorded.

Ophthalmological examination included refractometry, biomi-croscopy, gonioscopy, Goldmann applanation tonometry and binocular ophthalmoscopy. No glaucomatous cupping was ob-served in evaluation of the optic nerve head with the Volk lens. Visual fields were tested with the Octopus G1 program (Interzeag,Switzerland) and the results were normal.

Corneal endothelial cells were examined on the central cornea,using Keeler-Konan contact specular microscopy (Konan SP 5500).The endothelial cells were videotaped and analyzed according to theKonan cell analysis system KC-87 A, which determines the meancell density (cells/mm2), the smallest and largest cell sizes, the meancell size, the coefficient of variation in cell area (an objective mea-sure of polymegethism), and the frequency of hexagonal cells (anindex of pleomorphism). Tracing and digitization was performed byone examiner (K.V.) unaware of the exfoliation status of the eye.

The central corneal thickness (CCT) was measured with an ul-trasonic pachymeter (Humphrey Ultrasonic Pachometer). After adrop of anesthetic had been instilled into the lower cul-de-sac, theprobe tip of the pachymeter was held perpendicularly on the cor-nea and centered over the undilated pupil. Six to seven consecu-tive measurements of the CCT were recorded. To avoid falsely el-evated values by measurement slightly off-center, where the cor-nea is thicker, only the four lowest measurements were averaged.

Statistical methods

The t-test for dependent series was used to compare differencesbetween the fellow eyes. Pearson’s correlation coefficient was cal-culated for correlation analyses. All the variables were tested us-ing multiple linear regression analysis. A value of P<0.05 was se-lected as denoting statistical significance of differences.

All the patients gave their informed consent prior to their inclu-sion in the study. The study was performed in accordance with theethical standards laid down in the 1964 Declaration of Helsinki.

Results

Exfoliation material was not found on the corneal endo-thelium in any eye. A few pigment granules were found

on the corneal endothelium in 13/40 (33%) of the E andin 2/40 (5%) of the fellow NE eyes.

Central corneal thickness

CCT values were significantly higher in the E than in thefellow NE eyes (0.528±0.030, range 0.461–0.591 mm,vs 0.523±0.032, range 0.453–0.591 mm, P<0.01) (Fig. 1,Table 1).

Intraocular pressure

The E eyes had higher IOP than the fellow NE eyes(15.7±3.6 mmHg vs 14.4±2.9 mmHg, P<0.001). WhenIOP was corrected, using reported correction values forcorneal thickness [6], the E eyes still had higher IOPthan the fellow NE eyes (15.1±4.4 vs 14.2±3.7 mmHg,P<0.05) (Table 1).

Endothelial cell density (cells/mm2)

The mean endothelial cell density for the E eyes was2779±540 cells/mm2 (range 1668–3972 cells/mm2) and

660

Table 1 Central corneal thick-ness (CCT), intraocular pres-sure (IOP), IOP corrected ac-cording to CCT and endothelialcells in exfoliative and non-exfoliative eyes (n=40)

Parameter Exfoliative eyes Non-exfoliative eyes P(mean + SD) (mean + SD)

CCT (mm) 0.528±0.030 0.523±0.032 <0.01IOP (mmHg) 15.7±3.6 14.4±2.9 <0.001Corrected IOP (mmHg) 15.1±4.4 14.2±3.7 <0.05Cell density (cells/mm2) 2779±540 2870±386 NSCoefficient of variation in cell size 0.25±0.03 0.26±0.03 NSHexagonal cells (%) 80.5±6.5 82.0±5.0 NS

Fig. 1 Central corneal thickness in exfoliative and fellow non-ex-foliative eyes

for the NE eyes, 2870±386 cells/mm2 (range 2069–3619 cells/mm2) (Table 1).The difference was not signif-icant (P=0.068). The endothelial cell density was lower(13–1018 cells/mm2) in the E than in the fellow NE eyein 25 patients, the same in 1 patient and higher in 14 pa-tients (84–667 cells/mm2). The endothelial cell densitywas, on average, 3.6±11.4% lower in the E eyes.

Variation in cell size (polymegethism)

The E and the NE eyes did not differ in the coefficient ofvariation in cell size (0.25±0.03 vs 0.26±0.03) (Table 1).

Percentage of hexagonal cells (pleomorphism)

The E and NE eyes did not differ in the frequency ofhexagonal cells (80.5±6.5 vs 82.0±5.0%) (Table 1).

Correlations

IOP did not correlate with age or with corneal parame-ters (CCT, cell density, variation in cell size, percentageof hexagonal cells) in the E or NE eyes. Age correlatedpositively only with variation in cell size in the NE eyes(P<0.05). There was a negative correlation between thevariation in cell size and the frequency of hexagonalcells in the E (P<0.01) and the NE (P<0.05) eyes. Thefrequency of hexagonal cells correlated positively withthe endothelial cell density in the E eyes (P<0.01).

Regression analysis

In multiple linear regression analysis with the enter pro-cedure the difference in CCT between the fellow eyescould not be attributed to age, sex, the difference in IOPor the difference in endothelial cell density.

Discussion

Forty non-glaucomatous normotensive patients with unilateral EXS were selected for the present study to in-vestigate the role of EXS in changes in CCT and in the corneal endothelium. The non-glaucomatous, non-exfoliative fellow eyes served as controls. Any effects ofage, sex, environment, systemic diseases and systemicmedication were ruled out by comparing the fellow eyes.

The etiology of EXS is unknown. There is disagree-ment as to whether EXS is ever unilateral or merelymanifests asymmetrically [14, 30]. Immunohistochemi-cal studies have revealed that, in unilateral cases, there isa similar microangiopathy of the anterior uvea in the

E eyes and the fellow eyes which showed no exfoliationeven by histopathology[14]. When first clinically detect-ed, EXS is more often unilateral than bilateral [15, 32].Conversion rates to bilateral disease range from 14% to41% within 5 years [8,15]. A low cumulative probabilityof 6.8% at 5 years and of 16.8% at 10 years has alsobeen calculated [10]. However, even with long follow-up, a number of patients continue to show exfoliation de-posits in one eye only. Hence the term clinically unilater-al EXS has been adopted.

Clinical signs related to dispersion of pigment fromthe iris are known to appear first, before exfoliation ma-terial can be detected by biomicroscopy [26, 30]. Devel-opment of the syndrome from the first pigmentarychanges to the full-scale exfoliation picture is supposedto take 5–10 years [12]. In the present study the criterionfor unilateral EXS was biomicroscopically detectable ex-foliation material on the lens capsule or at the pupillaryborder. It is possible that some of the NE eyes represent-ed preclinical stages of EXS even though they did noteven reveal signs related to pigment dispersion.

The E eyes in this study represented early stages ofEXS. Exfoliation material could not be detected on thecorneal endothelium in any eye. Only a few pigmentgranules were seen on the corneal endothelium in 13 ofthe 40 E eyes and 2 of the 40 NE eyes. The eyes werenormotensive.

In this study the mean values for CCT (0.528 mm forthe E and 0.523 mm for the NE eyes) did not differ fromthose published earlier for the normal population. Sever-al studies agree on an average CCT of about 0.520 mmin normal eyes ( Lowe [18]: 0.517 mm; Hansen [9]:0.520 mm for right and 0.524 mm for left eyes; Mishimaand Hedbys [19]: 0.518 mm), although higher valueshave been measured in normal subjects (Herndon et al.[11]: 0.561 mm, Bron et al. [3]: 545 mm). CCT seems tobe an independent variable unrelated to other ocular di-mensions (refraction, anterior corneal curvature, axiallength) and sex [7, 9, 18, 25]. The CCT value has beenshown to be independent of age [3, 9, 18] or to declinewith age [25]. In this study, we found no correlation be-tween age and CCT in either E or NE eyes.

In the present study the E eyes had significantly high-er values for CCT than the fellow eyes in pairwise com-parisons. This finding is in accord with that of Ehlers et al. [7], who found CCT to be higher (although not statistically significantly so) in a group of eyes with untreated EXS (0.539±0.07 mm for the right and0.538±0.07 mm for the left eyes) than in a group of nor-mal eyes. In the present study the finding cannot be ex-plained by a side difference (exfoliation was present in20 right and 20 left eyes), though in one study the leftcorneas were found to be thicker than the right corneasin a normal population [25]. The finding may be ex-plained by a very early and subtle decompensation of thebarrier function of the endothelial cells or may reflect

661

structural changes in the corneal stoma in EXS [31]. Inaddition, a real exfoliation keratopathy has been de-scribed which may result in diffuse corneal endothelialdecompensation and is based on observations in sevencorneal specimens of patients with EXS (with or withoutglaucoma) and accumulations of retrocorneal exfoliationmaterial [23, 24, 29].

Corneal thickness is a factor of clinical importancewhen evaluating IOP. It is reduced in low-tension glau-coma, increased in ocular hypertension and normal ineyes with primary open-angle glaucoma [5, 7, 11]. MeanCCT has also been found to be positively correlated withapplanation pressure, even in normotensive subjects [3,6, 7, 9, 11]. We were unable to confirm this correlationin our normotensive subjects. In Bron’s study [3], thiscorrelation was also poor, suggesting that other factorsare involved.

Ehlers et al. [6] verified that, with a normal cornealthickness of about 0.520 mm, the Goldmann applanationtonometer gives a correct value for IOP. When the thick-ness is below normal the applanation reading is too low,and when it is above normal the reading is too high. Ac-cording to the correction table [6], applanation tonome-try is overestimated or underestimated by approximately 5 mmHg for every 0.070 mm change in CCT. Whitacre et al. also noticed that thin corneas produce underestima-tions of IOP by as much as 4.9 mmHg, whereas thickcorneas yield overestimations by as much as 6.8 mmHg[35].

The finding that IOP, on average, is higher in E thanNE eyes [8, 15, 27] was confirmed in this study. Also,when we corrected the IOP readings according to CCTand used the values for correction given by Ehlers et al.[6], the E eyes had higher IOP than the fellow eyes, butthe difference in IOP between the paired eyes was small-er. Thus, in the early stages of EXS, the higher IOP read-ings in E eyes than in fellow NE eyes seem only partlydue to the increased CCT in the E eyes.

The number of endothelial cells decreases with age by0.3–0.6% per year in adulthood [2, 22] with gradual in-creases in polymegethism and pleomorphism [2]. Theendothelial cell density is approximately 2690 cells/mm2

in a 61- to 70-year-old and 2400 cells/mm2 in a 71- to80-year-old subject [22]. Endothelial cell densities forthe E eyes (2826 cells/mm2) and the fellow NE eyes(2879 cells/mm2) in this study (mean age 68.4 years)were within normal limits.

In this study (n=40) the difference in endothelial celldensity between the paired eyes did not reach the levelof significance (P=0.068); nor did it do so in a study(n=17) by Miyake et al. [20]. Brooks et al. [4], however,found the endothelial cell count to be lower in affectedeyes than in fellow non-affected eyes in unilateral EXS,whether with (n=19) or without (n=12) a rise in IOP. Inthe same way, in patients (n=27) with unilateral capsularglaucoma Vannas et al. [34] found the mean corneal en-

dothelial cell density to be 5.2% lower in capsular glau-comatous eyes than in their fellow normotensive eyes.Endothelial cell density has also been shown to be9.9–10.5% lower in eyes with EXS than in age-matchedcontrol eyes that were otherwise healthy but had senilecataract [20, 36, 37].

We found no correlations between IOP and cornealendothelial parameters in these normotensive subjects.Similarly, in patients with unilateral capsular glaucomathe endothelial cell density did not correlate with theIOP, the difference in IOP between the fellow eyes, thevisual field defect or the duration of glaucoma [34]. Inone study, however, a rise in IOP accentuated the endo-thelial changes, although it was not necessary for the de-velopment of the changes [4].

In this study the E eyes did not have qualitativelypoorer endothelial cells than the fellow eyes. This find-ing is in accord with the study of Wirbelauer et al. [36,37], in which no qualitative differences between the Eeyes and normal control eyes, and no increased endothe-lial cell loss or differences in endothelial repair mecha-nisms, were found after a cataract operation in the Eeyes. In contrast, Brooks et al. [4] found mild pleomor-phism and polymegethism in affected eyes comparedwith their unaffected fellow eyes. In Brook’s study [4]the NE fellow eyes were regular, as in our study, but Miyake et al. [20] found pleomorphism and polymegeth-ism not only in the E eyes but also in the fellow NE eyescompared with normal control eyes. In the present studythe proportions of hexagonal cells in the E eyes (81%)and the NE eyes (82%) were higher than in earlier stud-ies (Miyake: 57% and 55% [20]; Wirbelauer: 61% and61% [37]), and the proportions of eyes showing polyme-gethism were lower (25% and 26%) in both the E andthe NE eyes than in earlier studies (Miyake: 34% and34% [20]; Wirbelauer: 36% and 35% [37].

In this study, in which the patients exhibited early stag-es of unilateral exfoliation, the E eyes did not show quanti-tative or qualitative morphological changes in corneal en-dothelial cells. The E eyes had significantly higher CCT,which may be a sign of decompensation of the barrierfunction of the endothelial cells or may reflect structuralchanges in the corneal stoma [31]. According to the pres-ent study, quantitative or qualitative morphological chang-es in the corneal endothelium seem not to be an early orconstant finding in EXS. A real exfoliation keratopathyseems to be a feature of advanced EXS. The E eyes hadhigher IOP than the fellow eyes, even after correcting theIOP values according to the CCT. Thus, in the early stagesof EXS, the higher IOP readings in E eyes than in NE eyesseem only partly due to increased CCT in the E eyes.

Acknowledgements This study was supported by a grant fromPaulo Säätiö (the Paulo Foundation) and Silmäsäätiö (the FinnishEye Foundation). The authors wish to thank Professor ChristinaRaitta and Professor Ahti Tarkkanen for their help in revising themanuscript.

662

663

References

1. Baba H (1982) Investigation of thepathogenesis of glaucoma capsularewith special reference of alfa 1 Lp andCp in aqueous humor. Graefe’s ArchClin Exp Ophthalmol 218: 283–286

2. Bourne WM, Nelson LR, Hodge DO(1997) Central corneal endothelial cellchanges over a ten-year period. InvestOphthalmol Vis Sci 38:779–782

3. Bron AM, Creuzot-Garcher C, Goudeau-Boutillon S, d’Athis P (1999)Falsely elevated intraocular pressuredue to increased central corneal thick-ness. Graefe´s Arch Clin ExpOphthalmol 237: 220–224

4. Brooks AMV, Robertson IF, Grant G,Gillies WE (1987) Progressive cornealendothelial cell changes in anteriorsegment disease. Austr N Z JOphthalmol 15: 71–78

5. Copt RP, Thomas R, Mermoud A(1999) Corneal thickness in ocular hy-pertension, primary open-angle glauco-ma, and normal tension glaucoma.Arch Ophthalmol 117: 14–16

6. Ehlers N, Bramsen T, Sperling S(1975) Applanation tonometry and central corneal thickness. Acta Ophthalmol (Copenh) 53: 34–43

7. Ehlers N, Hansen FK, Aasved H(1975) Biometric correlations of corne-al thickness. Acta Ophthalmol(Copenh) 53: 652–659

8. Hansen E, Sellevold OJ (1969) Pseudo-exfoliation of the lens capsule. II. De-velopment of the exfoliation syndrome.Acta Ophthalmol (Copenh) 47:161–173

9. Hansen FK (1971) A clinical study ofthe normal human central cornealthickness. Acta Ophthalmol (Copenh)49: 82–89

10. Henry JC, Krupin T, Schmitt M, Lauffer J, Miller E, Ewing MQ, ScheieHG (1987) Long-term follow-up ofpseudoexfoliation and the developmentof elevated intraocular pressure. Ophthalmology 94: 545–552

11. Herndon LW, Choudhri SA, Cox T,Damji KF, Shields MB, Allingham RR(1997) Central corneal thickness innormal, glaucomatous, and ocular hy-pertensive eyes. Arch Ophthalmol 115:1137–1141

12. Jerndal T (1986) Open angle glaucomaand the pseudo-exfoliation syndrome.In: Cairns JE (ed) Glaucoma, vol II.Grune & Stratton, London, pp 661–677

13. Johnson DH, Brubaker F (1982) Dy-namics of aqueous humor in the syn-drome of exfoliation with glaucoma.Am J Ophthalmol 93: 629–634

14. Kivelä T, Hietanen J, Uusitalo M(1997) Autopsy of clinically unilateralexfoliation syndrome. InvestOphthalmol Vis Sci 38: 2008–2015

15. Klemetti A (1988) Intraocular pressurein exfoliation syndrome. ActaOphthalmol (Copenh) 66 [Suppl 184]:54–58

16. Kühle M, Vinores SA, Mahlow J,Green WR (1996) Blood-aqueous bar-rier in exfoliation syndrome: evalua-tion by immunohistochemical stainingof endogenous albumin. Graefe’s ArchClin Exp Ophthalmol 234: 12–18

17. Laatikainen L (1971) Fluorescein an-giographic studies of the peripapillaryand perilimbal regions in simple, cap-sular and low-tension glaucoma. ActaOphthalmol (Copenh) Suppl 111: 1–83

18. Lowe RD (1969) Central corneal thick-ness. Ocular correlations in normaleyes and those with primary angle-closure glaucoma. Br J Ophthalmol 53:824–826

19. Mishima S, Hedbys BO (1968) Mea-surement of corneal thickness with theHaag-Streit pachometer. ArchOphthalmol 80: 710–713

20. Miyake K, Matsuda M, Inaba M (1989)Corneal endothelial changes in pseudo-exfoliation syndrome. Am J Ophthal-mol 108: 49–52

21. Morrison JC, Green R (1988) Light mi-croscopy of the exfoliation syndrome.Acta Ophthalmol (Copenh) 66 [Suppl184]: 5–27

22. Møller-Pedersen T (1997) A compara-tive study of human corneal keratocyteand endothelial cell density during ag-ing. Cornea 16: 333–338

23. Nauman GOH (1998) Corneal thick-ness and applanation pressure. ArchOphthalmol 116: 701

24. Naumann GOH, Schlötzer-SchrehardtU (1994) Corneal endothelial involve-ment in pseudoexfoliation syndrome.Arch Ophthalmol 112: 297–298

25. Olsen T, Ehlers N (1984) The thicknessof the human cornea as determined bya specular method. Acta Ophthalmol(Copenh) 62: 859–871

26. Prince AM, Ritch R (1986) Clinicalsigns of the pseudoexfoliation syn-drome . Ophthalmology 93: 803–807

27. Puska P, Raitta C (1992) Exfoliationsyndrome as a risk factor for optic discchanges in nonglaucomatous eyes.Graefe’s Arch Clin Exp Ophthalmol230: 501–504

28. Schlötzer-Schrehardt U, Küchle M,Naumann GOH (1991) Electronmicro-scopic identification of pseudoexfolia-tion material in extrabulbar tissue.Arch Ophthalmol 109: 565–570

29. Schlötzer-Schrehardt U, Dörfler S,Naumann GOH (1993) Corneal endo-thelial involvement in pseudoexfolia-tion syndrome. Arch Ophthalmol 111:666–674

30. Schlötzer-Schrehardt U, Küchle M,Dörfler S, Naumann GOH (1993)Pseudoexfoliative material in the eye-lid skin of pseudoexfoliation-suspectpatients: clinico-histopathological cor-relation. Ger J Ophthalmol 2: 51–60

31. Schlötzer-Schrehardt U, Mark von derK, Sakai LY, Naumann GOH (1997)Increased extracellular deposition of fi-brillin-containing fibrils in pseudoexfo-liation syndrome. Invest OphthalmolVis Sci 38: 970–984

32. Tarkkanen A (1962) Pseudoexfoliationof the lens capsule. A clinical study of418 patients with special reference toglaucoma, cataract, and changes of thevitreous. Acta Ophthalmol (Copenh)Suppl 71: 1–98

33. Vannas A (1969) Fluorescein angiogra-phy of the vessels of the iris in pseudo-exfoliation of the lens capsule, capsularglaucoma and some other forms ofglaucoma. Acta Ophthalmol (Copenh)Suppl 105: 1–75

34. Vannas A, Setälä K, Ruusuvaara P(1977) Endothelial cells in capsularglaucoma. Acta Ophthalmol (Copenh)55: 951–958

35. Whitacre MM, Stein RA, Hassanein K(1993) The effect of corneal thicknesson applanation tonometry. Am JOphthalmol 115: 592–596

36. Wirbelauer C, Anders N, Pham DT,Holschbach A, Wollensak J (1997)Frühpostoperativer Endothelzellverlustnach korneoskleralem Tunnelschitt undPhakoemulsifikation bei Pseudoexfoli-ationssyndrom. Ophthalmologe 94:332–336

37. Wirbelauer C, Anders N, Pham DT,Wollensak J (1998) Corneal endothelialcell changes in pseudoexfoliation syn-drome after cataract surgery. ArchOphthalmol 116: 145–149