Introduction

In pseudoexfoliation syndrome (PES),there is deposition and accumulationof a granular microfibrillar materialthroughout the anterior segment ofthe eye and throughout the body inother organ systems. The frequency of

this condition varies in different coun-tries (Drolsum et al. 2007). Sixteenper cent of the patients who werereferred for cataract surgery to theEye Department in Oslo had PES(Drolsum et al. 1993a, 1998).

Pseudoexfoliation syndrome is asso-ciated with glaucoma (Drolsum et al.

1993a). There is an increased risk ofcomplicated cataract surgery becauseof atrophy of the iris and pupillary ruff,with insufficient mydrasis, weak zonu-lae and increased risk of capsule ⁄ zonu-lae rupture with vitreous loss (Drolsumet al. 1993a, 1998; Shingleton et al.2003). Increased rates of inflammation,secondary cataract and late dislocationof the intraocular lens (IOL) withinthe capsular bag have been reportedpostoperatively (Drolsum et al. 1993b;Kuchle et al. 1997; Gimbel et al. 2005).

Specular and electron microscopicstudies have revealed both quantita-tive and qualitative corneal endothe-lial cell changes in eyes with PES. InPES eyes, not operated for cataract,several authors have found a lowerendothelial cell density (ECD).Changes in the percentage of hexago-nal cells (an index of pleomorphism)and the coefficient of variation of cellsize (a measure of polymegatism) havebeen shown when compared to eyeswithout PES (Miyake et al. 1989;Stefaniotou et al. 1992; Wirbelaueret al. 1997; Naumann & Schlotzer-Schrehardt 2000; Inoue et al. 2003;Wali et al. 2009). However, there arefew studies on ECD after cataract sur-gery in PES eyes. In these studies, thesample sizes are relatively small andthe endothelial measurements wereperformed shortly following cataractsurgery (Wirbelauer et al. 1997, 1998;Kaljurand & Teesalu 2007). It hasbeen postulated that in PES eyes,there might be a higher risk of

Corneal endothelial cells6–7 years following cataractsurgery in patients withpseudoexfoliation syndrome

Atle E. Østern1 and Liv Drolsum2

1Eye Department, Oslo University Hospital, Oslo, Norway2Eye Department, Oslo University Hospital and University of Oslo, Oslo, Norway

ABSTRACT.

Purpose: To assess the condition of the corneal endothelium an extended per-

iod after cataract surgery in eyes with and without pseudoexfoliation syndrome

(PES).

Methods: Forty-six patients with PES who underwent cataract surgery in the

Eye Department, Oslo University Hospital, in 2001 and 2002 were enrolled

and compared to 101 matched controls without PES who had surgery in the

same period. They were re-examined 6–7 years following surgery with mea-

surements taken of corneal endothelial cell density (ECD), pleomorphism,

polymegathism and corneal thickness.

Results: Mean ECD was 2024 ± 371 cells ⁄mm2 in eyes with PES and

2144 ± 365 cells ⁄mm2 in eyes without PES. The difference was not statisti-

cally significant. No significant difference in polymegathism and pleomorphism

was noted. Mean corneal thickness was 543 and 547 lm in eyes with and with-

out PES, respectively (not statistically significant). The presence of glaucoma

in pseudoexfoliative eyes was not associated with endothelial cell changes.

Conclusion: Six to 7 years following cataract surgery, no statistically signifi-

cant differences were established in ECD, pleomorphism, polymegathism and

corneal thickness in eyes with and without PES. No clinical signs of corneal

decompensation were noted amongst the participants.

Key words: cataract surgery – corneal endothelial cell density – corneal thickness – glaucoma –

pseudoexfoliation syndrome

Acta Ophthalmol.ª 2010 The Authors

Acta Ophthalmologica ª 2010 Acta Ophthalmologica Scandinavica Foundation

doi: 10.1111/j.1755-3768.2010.02012.x

Acta Ophthalmologica 2010

1

progressive decline in ECD after cata-ract surgery because of low gradeinflammation (Wirbelauer et al. 1998).

In our cataract department, we havea relatively high proportion of patientswith PES. Our clinical impression wasthat these patients are no more proneto corneal decompensation over anextended time period when comparedto patients without PES. The purposeof this study is to compare thecorneas of patients with and withoutPES in terms of ECD, endothelialmorphology and corneal thickness6–7 years following cataract surgery.

Material

A study was performed comparingtwo age- and gender-matched groups,one with PES, the other without. Bothgroups were operated upon betweenJune 2001 and December 2002 andre-examined in 2008.

A total of 1654 eyes from 1193patients were operated for cataractin the defined period in 2001 and2002. Pseudoexfoliation syndromewas recognized in 179 of the eyes(10.8%). Four hundred and twenty-three (35.4%) patients died before2008. Of the remaining 770 patients in2008, 96 had PES in one or both eyesat their initial preoperative examina-tion. Exclusion criteria were as fol-lows: age below 60 years at the timeof surgery, intraocular surgery or eyetrauma after cataract extraction orinability to carry out tests for physicalor mental reasons. This resulted in 22patients being excluded. Of theremaining 74 patients, 19 (25.7%) hadmoved from the Oslo area or theirpresent whereabouts were unknown.Four patients (5.4%) declined toparticipate. Fifty-one patients withPES were thus re-examined in 2008,together with 102 controls, matchedfor age and gender. Five of thepatients with PES and one in the con-trol group were unable to completethe endothelial cell count and wereconsequently excluded. In total, 46PES eyes and 101 controls wereenrolled. Only the first operated eyefrom the defined period was examined.

The records from 2001 to 2002 werereviewed. Before cataract extraction,the eyes had been dilated and thepseudoexfoliative material was notedusing slitlamp biomicroscopy. Phaco-emulsification was used in all cases.

In two patients with and 10 patientswithout PES, the IOL had beenplaced in the ciliary sulcus. In theremaining eyes, the IOL had beenimplanted in the capsular bag.Posterior capsular tear without vitre-ous loss had occured in two eyeswithout PES and posterior capsulartear with vitreous loss in one eyewith PES.

In 2008, we assessed the condition ofthe corneal endothelium of the enrolledpatients by means of several investiga-tive methods. Pachymetry values forcorneal thickness were obtained byPentacam HR (Oculus OptikgerateGmbH, Wetzlar, Germany). For mea-suring the corneal endothelial cells, anendothelial cell counter, ConfoScan 4Confocal Microscope (Nidek Techno-logies, Padova, Italy), was employed.The built-in, fully automated, noncon-tact, confocal endothelial microscopewas used, which is designed specificallyfor analysis of the endothelial cell layerwith less disturbance by corneal haze.The instrument provides a large mea-surement area of up to 1000 cells ⁄exam, which ensures reliability andaccuracy. Endothelial cell density,polymegathism and pleomorphismwere measured in the centre of the cor-nea. The mean ECD of each patientwas based upon three calculations fromthe recorded images. If any of the val-ues diverged by more than 5%, theprocedure was repeated two moretimes. The value that differed the mostwas excluded before calculation of thedensity.

A complete ophthalmological exami-nation including slitlamp examinationof the cornea was performed. Clinicalsigns of corneal oedema were noted.A diagnosis of glaucoma was basedupon measurement of the intraocularpressure, optic nerve cupping and visualfield measurement, when indicated.

Informed written consent wasobtained from all patients. The study

was approved by the regional commit-tee for medical and health researchethics.

Statistics

Sample size calculations demonstratedthat with 50 eyes with PES and 100 eyeswithout PES and a 5% significancelevel, the study would have a power of80% to detect a difference of 10% inECD. An independent t-test was usedwhen comparing mean endothelial cellchanges and corneal thickness in thetwo groups. Results are expressed asmean ± 1 standard deviation. A sig-nificance level of 5% was used through-out. The statistical analysis wasperformed using spss version 15.0(SPSS Inc., Chicago, Illinois, USA).

Results

Characteristics of age and gender areshown in Table 1. Mean age of thepatients in the PES group was82.1 ± 4.9 years (range 72–89 years)and for the patients in the non-PESgroup 82.3 ± 5.1 years (range 71–93 years).

Our main results are presented inTable 2. We found that there was atendency for lower ECD in PES eyescompared to non-PES eyes. The dif-ference was, however, not statisticallysignificant (p = 0.07). Furthermore,no statistically significant differencesin pleomorphism and polymegatismwere found between the groups.Although central corneal thickness inboth the thinnest and central partswas slightly lower in eyes with PES,the difference was not statistically sig-nificant.

Endothelial cell density, polyme-gathism and pleomorphism in eyeswith and without glaucoma are shownin Table 3. There were no significantdifferences in these measurementsbetween eyes with and without PES.

Table 1. Age and gender distribution in the PES group and controls.

Eyes with PES

(n = 46)

Eyes without

PES (n = 101)

Patient age ± 1 SD

(range min–max)

82.1 ± 4.9 years

(range 72–89)

82.3 ± 5.1 years

(range 71–93)

Patient gender

Male (%) 12 (26.1) 28 (27.7)

Female (%) 34 (73.9) 73 (72.3)

PES = pseudoexfoliation syndrome; SD = standard deviation.

Acta Ophthalmologica 2010

2

None of the patients in our study,with or without PES, had clinicalsigns of corneal endothelial decom-pensation.

Discussion

To the best of our knowledge, thereare no previous reports on the state ofthe cornea and the corneal endothe-lium in PES a substantial period fol-lowing cataract surgery. In our studyof a relatively large group of patients6–7 years after surgery, we found nostatistical significant differences inECD, polymegathism, pleomorphismor corneal thickness comparing PESwith non-PES eyes.

Endothelial cell density has, how-ever, been reported to be lower inmost studies comparing PES eyesbefore cataract surgery with non-PESeyes (Wirbelauer et al. 1997, 1998;Inoue et al. 2003). It has been pro-posed that these endothelial changesseen in PES eyes are induced byaltered composition and increasedflare intensity of the aqueous humour,caused by breakdown of the blood–aqueous barrier (Schlotzer-Schrehardtet al. 1992). Electron microscopy has

shown large aggregations of typicalpseudoexfoliation (PE) materialadhering to the corneal endothelium(Naumann & Schlotzer-Schrehardt2000). In fact, it has been proposedthat the PE material is formed byfocally detached endothelial cells withhypertrophied secretory organelles(Schlotzer-Schrehardt et al. 1993;Naumann & Schlotzer-Schrehardt2000). Descemet’s membrane appearsthickened, with irregular excrescencesand diffuse melanin deposition.According to these studies, an abnor-mal extracellular matrix is synthesized,composed of basement membranematerial, microfibrils and PE fibres.These changes may lead to early cor-neal endothelial decompensation.

Therefore, one might anticipate thatthe ECD difference between eyes withand without PES would continue toincrease, with or without surgery.However, our results may suggestotherwise, given that ECD before sur-gery did not differ greatly from whathas been observed in other studies.Prior to surgery, there is mechanicalfriction between the iris and the lens,which could lead to microfibrillarmaterial being rubbed off of the sur-

face of the anterior lens capsule (Ritch1994). Some researchers have suggestedthat a proportion of the subsequentlyaccumulated fibrillar eosinophilicmaterial sticks to the corneal endothe-lium because of its adhering capabili-ties (Eagle et al. 1979; Ringvold1994). If this is the case, some of thePE fibrils observed on the surface ofthe endothelial cells may originallycome from the lens. In addition, thefragile pupillary margin of the irismay cause increased pigment disper-sion in eyes with PES (Prince & Ritch1986). Melanin granules, derived fromiris pigment, are then phagocytosedby endothelial cells (Naumann & Sch-lotzer-Schrehardt 2000). One maytherefore speculate that removing thelens could lead to less negative influ-ence on the corneal endothelium inPES eyes. If so, this hypothesis mightexplain our finding that the ECD, themorphology of the endothelial cellsand corneal thickness were quite com-parable in PES and non-PES eyes inthis study.

In our study, mean ECD was aslow as 2024 and 2144 cells ⁄mm2 ineyes with and without PES, respec-tively. The low density in both groupsis explained by the high mean age ofthe patients, likely in combinationwith previous cataract surgery. Physi-ological reduction in endothelial cellsbecause of ageing has been estimatedto be 0.5–0.6% per year (Bourne et al.1994). When it comes to the effect ofphacoemulsification, there are only afew studies conducted on PES eyes,all during the early postoperative per-iod. In one study undertaken 1 monthfollowing surgery, reduction in ECDin the PES group (n = 27) was 18.1%compared to 11.6% in the non-PESgroup (n = 26) with a weak signifi-cance (p = 0.06) (Kaljurand & Tee-salu 2007). In another study,

Table 2. A comparison of CCT, ECD, endothelial polymegathism and pleomorphism in eyes with and without PES.

Eyes with PES

(n = 46)

Eyes without PES

(n = 101)

pMean ± 1 SD Range min ⁄max Mean ± 1 SD Range min ⁄max

ECD 2024 ± 371 1328–2714 2144 ± 365 1301–3398 0.07

Polymegathism 0.440 ± 0.130 0.290–1.08 0.445 ± 0.112 0.230–0.800 0.99

Pleomorphism % 46.4 ± 10.1 26–65 46.3 ± 10.9 23–75 0.99

CCT thinnest part in lm 543 ± 40 474–687 547 ± 35 424–646 0.56

CCT central part in lm 550 ± 42 419–698 554 ± 33 434–649 0.74

ECD = endothelial cell density; PES = pseudoexfoliation syndrome; SD = standard deviation; CCT = central corneal thickness.

Table 3. ECD, polymegathism and pleomorphism in PES and non-PES eyes with and without

glaucoma.

Eyes without glaucoma Eyes with glaucoma

pMean ± 1 SD Mean ± 1 SD

ECD

PES eyes 2029 ± 405 (n = 30) 2012 ± 321 (n = 15) 0.68

Non-PES eyes 2154 ± 371 (n = 88) 1949 ± 251 (n = 9) 0.78

Polymegathism

PES eyes 0.452 ± 0.147 0.419 ± 0.096 0.43

Non-PES eyes 0.442 ± 0.113 0.466 ± 0.113 0.56

Pleomorphism %

PES eyes 44.8 ± 9.4 48.6 ± 11.3 0.24

Non-PES eyes 46.0 ± 10.9 49.8 ± 12.5 0.32

ECD = endothelial cell density; PES = pseudoexfoliation syndrome; SD = standard devia-

tion.

Acta Ophthalmologica 2010

3

postoperative endothelial cell loss6 months after cataract surgery was11.1% in the PES group (n = 25) and10.3% in the control group (n = 25).The difference was not significant(Wirbelauer et al. 1998).

We found no reduction in ECDaccording to glaucoma status. Neitherwere there significant differences forpolymegathism or pleomorphism. Datafrom one study suggest a lower ECDin patients with glaucoma than inthose without glaucoma, however,with no significant differences forpleomorphism and polymegathism(Gagnon et al. 1997). Ocular hyper-tension and primary open-angle glau-coma had no adverse effect on ECDin another study (Korey et al. 1982).Some authors have reported a morepronounced endothelial cell deteriora-tion in eyes with both PES and glau-coma than in eyes with only glaucoma(Inoue et al. 2002; Wali et al. 2009).

A prospective study would be theideal way of evaluating possible futuredifferences in the condition of theendothelium in the two groups withand without PES. However, a pro-spective study would most likely nothave provided us with more reliabledata compared with our study.Because the mean age of patients withPES who are operated for cataract inour area is as high as 76 years, along-term prospective study of6–7 years would have meant signifi-cant problems with attrition and feasi-bility. With a high proportion lost tofollow-up, major difficulties withinterpretation of results can occur.

We used a noncontact endothelialmicroscope with a 20· probe to mea-sure endothelial corneal changes. Thealternative methods with a gel immer-sion 40· lens or Z ring would haveled to an increased inability to carryout the necessary tests because of thehigh mean age of our patients(82 years), likely without resulting inhigher reliability for ECD values.

To summarize, we found no signifi-cant difference in the ECD betweeneyes with and without PES a substan-tial period following cataract surgery.Our study included more patients thanmost other comparable studies. Therewere no cases of corneal oedema inthe PES-affected eyes. Our study indi-cates that besides the usual precau-tions which should be taken intoaccount because of the well-estab-

lished risks in PES eyes, there is noreason to refrain from surgery becauseof the cornea or for special follow-upafter uneventful phacoemulsification.

Acknowledgments

This study was supported by a grantfrom Oslo University Hospital, Nor-way and Arthur and Odd Clauson’slegate, Oslo, Norway.

ReferencesBourne WM, Nelson LR & Hodge DO

(1994): Continued endothelial cell loss ten

years after lens implantation. Ophthalmol-

ogy 101: 1014–1022.

Drolsum L, Haaskjold E & Davanger M

(1993a): Pseudoexfoliation syndrome and

extracapsular cataract extraction. Acta

Ophthalmol (Copenh) 71: 765–770.

Drolsum L, Haaskjold E & Davanger M

(1993b): Results and complications after

extracapsular cataract extraction in eyes

with pseudoexfoliation syndrome. Acta

Ophthalmol (Copenh) 71: 771–776.

Drolsum L, Haaskjold E & Sandvig K

(1998): Phacoemulsification in eyes with

pseudoexfoliation. J Cataract Refract Surg

24: 787–792.

Drolsum L, Ringvold A & Nicolaissen B

(2007): Cataract and glaucoma surgery in

pseudoexfoliation syndrome: a review. Acta

Ophthalmol Scand 85: 810–821.

Eagle RC Jr, Font RL & Fine BS (1979):

The basement membrane exfoliation syn-

drome. Arch Ophthalmol 97: 510–515.

Gagnon MM, Boisjoly HM, Brunette I, Cha-

rest M & Amyot M (1997): Corneal endo-

thelial cell density in glaucoma. Cornea 16:

314–318.

Gimbel HV, Condon GP, Kohnen T, Olson

RJ & Halkiadakis I (2005): Late in-the-bag

intraocular lens dislocation: incidence, pre-

vention, and management. J Cataract

Refract Surg 31: 2193–2204.

Inoue K, Tokuda Y, Inoue Y, Amano S,

Oshika T & Inoue J (2002): Corneal endo-

thelial cell morphology in patients undergo-

ing cataract surgery. Cornea 21: 360–363.

Inoue K, Okugawa K, Oshika T & Amano S

(2003): Morphological study of corneal

endothelium and corneal thickness in

pseudoexfoliation syndrome. Jpn J Oph-

thalmol 47: 235–239.

Kaljurand K & Teesalu P (2007): Exfoliation

syndrome as a risk factor for corneal endo-

thelial cell loss in cataract surgery. Ann

Ophthalmol (Skokie) 39: 327–333.

Korey M, Gieser D, Kass MA, Waltman SR,

Gordon M & Becker B (1982): Central cor-

neal endothelial cell density and central

corneal thickness in ocular hypertension

and primary open-angle glaucoma. Am J

Ophthalmol 94: 610–616.

Kuchle M, Amberg A, Martus P, Nguyen

NX & Naumann GO (1997): Pseudoexfoli-

ation syndrome and secondary cataract. Br

J Ophthalmol 81: 862–866.

Miyake K, Matsuda M & Inaba M (1989):

Corneal endothelial changes in pseudoexfo-

liation syndrome. Am J Ophthalmol 108:

49–52.

Naumann GO & Schlotzer-Schrehardt U

(2000): Keratopathy in pseudoexfoliation

syndrome as a cause of corneal endothelial

decompensation: a clinicopathologic study.

Ophthalmology 107: 1111–1124.

Prince AM & Ritch R (1986): Clinical signs

of the pseudoexfoliation syndrome. Oph-

thalmology 93: 803–807.

Ringvold A (1994): Corneal endothelial

involvement in pseudoexfoliation syn-

drome. Arch Ophthalmol 112: 297–298.

Ritch R (1994): Exfoliation syndrome and

occludable angles. Trans Am Ophthalmol

Soc 92: 845–944.

Schlotzer-Schrehardt UM, Koca MR, Nau-

mann GO & Volkholz H (1992): Pseudoex-

foliation syndrome. Ocular manifestation

of a systemic disorder?. Arch Ophthalmol

110: 1752–1756.

Schlotzer-Schrehardt UM, Dorfler S & Nau-

mann GO (1993): Corneal endothelial

involvement in pseudoexfoliation syn-

drome. Arch Ophthalmol 111: 666–674.

Shingleton BJ, Heltzer J & O’Donoghue MW

(2003): Outcomes of phacoemulsification in

patients with and without pseudoexfolia-

tion syndrome. J Cataract Refract Surg 29:

1080–1086.

Stefaniotou M, Kalogeropoulos C, Razis N

& Psilas K (1992): The cornea in exfolia-

tion syndrome. Doc Ophthalmol 80: 329–

333.

Wali UK, Bialasiewicz AA, Rizvi SG &

Al-Belushi H (2009): In vivo morphometry

of corneal endothelial cells in pseudoexfoli-

ation keratopathy with glaucoma and cata-

ract. Ophthalmic Res 41: 175–179.

Wirbelauer C, Anders N, Pham DT, Holsch-

bach A & Wollensak J (1997): [Early post-

operative endothelial cell loss after

corneoscleral tunnel incision and phaco-

emulsification in pseudoexfoliation syn-

drome]. Ophthalmologe 94: 332–336.

Wirbelauer C, Anders N, Pham DT & Wol-

lensak J (1998): Corneal endothelial cell

changes in pseudoexfoliation syndrome

after cataract surgery. Arch Ophthalmol

116: 145–149.

Received on June 7th, 2010.

Accepted on August 13th, 2010.

Correspondence:

Atle Einar Østern, MD

Eye Department

Oslo University Hospital (Ulleval)

0407 Oslo

Norway

Tel: + 47 230 15635

Fax: + 47 221 19989

Emails: atst@uus.no; aoestern@online.no

Acta Ophthalmologica 2010

4