ARTICLE

Factors related to corneal endothelialdamage after phacoemulsificationin eyes with occludable angles

Yu-Chieh Ko, MD, Catherine Jui-ling Liu, MD, Ling-Ing Lau, MD,Chih-Wei Wu, MD, Joe C. Chou, MD, Wen-Ming Hsu, MD

PURPOSE: To evaluate the influence of preoperative variables and early postoperative intraocularpressure (IOP) on the loss of corneal endothelial cells after phacoemulsification in eyes with occlud-able angles.

SETTING: Taipei Veterans General Hospital, Taipei, Taiwan.

METHODS: Sixty patients with occludable angles having phacoemulsification were prospectivelyenrolled. Corneal endothelial cell evaluation was conducted preoperatively and 3 monthspostoperatively.

RESULTS: Three months postoperatively, the mean corneal endothelial cell density decreased by14.5% G 25.8% (SD) (P < .001). Greater corneal endothelial cell loss was associated with shorteraxial length (AL) (P Z .008), steeper anterior corneal curvature (P Z .03), greater nuclear opales-cence (P Z .04), and higher IOP measured 4 to 8 hours after surgery (P Z .04) and the followingmorning (P Z .002). Multiple linear regression analysis identified AL and the IOP measured 4 to 8hours after surgery as the best predictors of postoperative corneal endothelial cell loss after adjustingfor nuclear opalescence and phacoemulsification time (R2 Z 0.40, P Z .001).

CONCLUSIONS: The corneal endothelial cell loss after phacoemulsification in eyes with occludableangles was associated with preoperative AL measurement and postoperative IOP within 24 hours.To minimize corneal endothelial cell damage, it is critical to avoid an IOP spike during the early post-operative period and to exercise extreme caution intraoperatively in eyes with an AL less than22.6 mm.

J Cataract Refract Surg 2008; 34:46–51 Q 2008 ASCRS and ESCRS

Eyes with occludable angles, characterized bya crowded anterior segment, are at risk for developingprimary angle-closure glaucoma (PACG), the leadingcause of glaucoma-blindness in Asia.1,2 Eyes at riskaccount for approximately 6% of the population inMongolia and Singapore.1 Although laser iridotomyeffectively relieves the pupillary block mechanismfor angle closure,3 a large and anteriorly positionedlens may result in residual angle closure and elevatedintraocular pressure (IOP) in postiridotomy eyes.2,4

Meanwhile, accumulating evidence indicates lens ex-traction effectivelywidens the drainage angle and con-trols IOP in eyes with occludable angles,4–7 althoughtransient elevation of postoperative IOP is common.5,7

Lens extraction with phacoemulsification is inevita-bly accompanied by loss of corneal endothelial cells,and eyes with crowded anterior segments may be

Q 2008 ASCRS and ESCRS

Published by Elsevier Inc.

46

more susceptible to this damage. Our pilot study(unpublished) found that the corneal endothelial cellloss 3 months after phacoemulsification was greater ineyeswithoccludableangles (28 eyes, 18.08%) than thosewith normal open angles (21 eyes, 4.99%) (P Z .027),while the preoperative corneal endothelial cell indices,nuclear opalescence, and phacoemulsification parame-ters did not differ between the 2 groups (unpublisheddata). Some studies report corneal decompensation af-ter uneventful cataract surgery in eyes with PACG.5,7

With the growing enthusiasm to treat primary angleclosure (PAC) andPACGwith phacoemulsification4,6,7

and the concern that eyeswithPACorPACGmayhaveless corneal endothelial cells than the age-matchednonglaucomatous eyes,8,9 it is important to identifyfactors predicting corneal endothelial cell loss afterphacoemulsification in these eyes at risk.

0886-3350/08/$dsee front matter

doi:10.1016/j.jcrs.2007.07.057

47CORNEAL ENDOTHELIUM AFTER PHACOEMULSIFICATION IN EYES WITH OCCLUDABLE ANGLES

This study aimed to identify the preoperative factorsassociated with changes in corneal endothelial cellcounts 3 months after phacoemulsification in eyeswith occludable angles. In addition, we evaluatedthe influence of early postoperative IOP on the cornealendothelium.

PATIENTS AND METHODS

This prospective interventional case series comprised 60 con-secutive Chinese patients with occludable angles havingphacoemulsification performed by the same surgeon(C.J.L.) in Taipei Veterans General Hospital, Taiwan,between July 2003 and October 2004. The chamber anglewas defined as occludable if 270 degrees or more of posteriortrabecular meshwork was not visible when gonioscopy wasperformed in the primary position with a Posner goniolensusing the smallest slitlamp beam in a dark room. The eyeswere further classified as PACG (32 eyes), PAC (9 eyes), orprimary angle-closure suspect (PACS) (14 eyes) accordingto the system proposed by Foster et al.10 Briefly, eyes withPACS have occludable angles but no peripheral anterior syn-echias, IOP elevation, or glaucomatous optic neuropathy.Eyes with PAC have occludable angles and established pe-ripheral anterior synechias or evidence of previous acute ep-isodes or elevated IOP, but no optic nerve changes. Primaryangle-closure glaucoma refers to eyeswith occludable anglesand established glaucomatous optic neuropathy.

To be enrolled, patients had to have a clear cornea withoutguttae and a patent peripheral iridotomy performed at least3months earlier. Eyeswith secondary angle closure, a historyof incisional surgery, or circumferential peripheral anteriorsynechias up to Schwalbe’s line were excluded. All partici-pants were examined as part of another study, whichinvestigated factors predicting postoperative IOP after pha-coemulsification.11 The institutional review board approvedthe study, and all patients provided written informedconsent.

Detailed ophthalmologic examinations, including Gold-mann tonometry, were performedwithin 2weeks before sur-gery and 3 months after surgery. In addition, IOP wasmeasured with the same noncontact tonometer (TopconCT60, Topcon Corp.) 4 to 8 hours and 20 to 24 hours (the fol-lowing morning) after phacoemulsification. The Lens Opac-ities Classification System III12 was used to grade nuclear

Accepted for publication July 31, 2007.

From the Department of Ophthalmology (Ko, Liu, Lau, Wu, Chou,Hsu), Taipei Veterans General Hospital, the National Yang MingUniversity School of Medicine (Ko, Liu, Lau, Chou, Hsu), and theInstitute of Clinical Medicine (Lau), National Yang-Ming University,Taipei, Taiwan.

No author has a financial or proprietary interest in any material ormethod mentioned.

Supported by research project VGH95-A100, Taipei VeteransGeneral Hospital, Taipei, Taiwan.

Corresponding author: Catherine Jui-ling Liu, MD, Department ofOphthalmology, Taipei Veterans General Hospital, 201, Shih-PaiRoad, Section 2, Taipei 11217, Taiwan. E-mail: jlliu@vghtpe.gov.tw.

J CATARACT REFRACT SUR

opalescence. The central corneal endotheliumwas evaluatedwith a noncontact specular microscope (SD-9000, Konan,Inc.). The central corneal thickness (CCT) was measuredwith an ultrasound pachymeter (DGH-550, DGH Technol-ogy, Inc.). A Tomey biometer and pachymeter (AL-1000, To-mey Corp.) was used to measure the anterior chamber depth(ACD), lens thickness, and axial length (AL). The mean of 3readings of each measurement was used for analysis. Thecorneal thickness was subtracted from the ACD measure-ment to obtain the actual ACD.

In all surgeries, a 3.2 mm temporal single-plane clear cor-neal incision was made with a side-port incision 80 degreesaway. After a manual continuous curvilinear capsulorhexiswas created with a bent needle and hydrodissection per-formed, phacoemulsification was performed using a Protegephacoemulsifier (DPX100, Storz Instruments). The divide-and-conquer and stop-and-chop techniques were used andwere admixed in some patients according to the intraopera-tive conditions. Sodium hyaluronate 3.0%–chondroitinsulfate 4.0% (Viscoat) was used as the ophthalmic viscosur-gical device. After cortical aspiration, an acrylic foldable in-traocular lens (AcrySof MA60BM, Alcon Laboratories, Inc.)was inserted and the corneal woundwas closedwith a single10-0 nylon suture. In cases with a small synechial pupil,Viscoat was used to separate the iris from the lens. Thephacoemulsification time was recorded.

In cases with postoperative IOP elevation, oral acetazol-amide or intravenous mannitol infusion was given. Preoper-ative glaucoma medications were maintained, except theprostaglandin analogue was replaced with oral acetazol-amide after surgery. Postoperative glaucoma medicationsbutwere adjusted based on the IOP. Topical steroid eyedropswere given postoperatively and tapered over 4 to 6 weeks.

Normality of data distribution was assessed by the 1-sam-ple Kolmogorov-Smirnov test. Comparisons between the 3diagnostic groups were made using the chi-square test or1-way analysis of variance. Postoperative changes in param-eters were evaluated by the paired Student t test orWilcoxonsigned rank test. The correlation between the postoperativechange in corneal endothelial cell density and the preopera-tive variables, phacoemulsification time, and early postoper-ative IOP was analyzed using the Pearson correlation orSpearman rank correlation. Variables with a P value lessthan 0.05 were included in a backward stepwisemultiple lin-ear regression analysis to identify independent predictors ofpostoperative corneal endothelial cell loss after the collinear-ity was assessed. The final model adopted was the onebelieved to adequately explain the data. All analyses wereconducted using SPSS statistical software (version 11.0.0,SPSS, Inc.).

RESULTS

Of the 60 eyes, 1 had vitreous loss and 4, incompletecorneal endothelial cell data; these 5 eyes were ex-cluded from analysis. The postoperative course wasuneventful in all patients; none developed severe ante-rior segment inflammation or required additional sur-gery during the study period.

The mean age of the patients was 73.9 years G 7.0(SD) (range 56 to 90 years), and 65.6% were men.Nine patients had a history of acute angle closurethat occurred more than 1 year previously.

G - VOL 34, JANUARY 2008

48 CORNEAL ENDOTHELIUM AFTER PHACOEMULSIFICATION IN EYES WITH OCCLUDABLE ANGLES

Preoperative gonioscopy showed that the posteriortrabecular meshwork remained invisible over 270 de-grees in 54 eyes (90%), even with a patent peripheraliridotomy. There were no statistically significant dif-ferences between the PACS, PAC, and PACG groupsor between eyes with and eyes without a history ofacute angle closure in the following parameters:preoperative biometric data, preoperative corneal en-dothelial cell indices, preoperative IOP, early postop-erative IOP, and postoperative changes in cornealendothelial cell indices. The ACD was statisticallysignificantly shallower in eyes with a history of acuteangle closure (PZ .017,Mann-Whitney test). Therefore,data from all patients were pooled for further analysis.The postoperative changes in clinical features areshown in Table 1.

Compared with preoperative values, the corneal en-dothelial cell density decreased by 14.5% G 25.8% 3months after phacoemulsification (P!.001). The corre-lations between the preoperative variables or earlypostoperative IOPs and the percentage of corneal en-dothelial cell loss 3 months after surgery are shownin Table 2. Eyes with steeper anterior corneal curva-ture, greater nuclear opalescence, shorter AL, andhigher IOP measured 4 to 8 hours postoperativelyand the following morning had greater corneal endo-thelial cell density loss 3 months postoperatively. Thephacoemulsification time was not associated with thepostoperative corneal endothelial cell loss (P Z .65).

In the final multiple regression model (Table 3),a shorter AL (P Z .002) and higher IOP 4 to 8 hourspostoperatively (P Z .001) were the best set of inde-pendent predictors of greater postoperative cornealendothelial cell loss, after adjusting for nuclear opales-cence and phacoemulsification time. Eyes with an ALshorter than the sample mean (22.6 mm) were morelikely than those with a longer AL to have more than20% of corneal endothelial cell loss, after adjusting

J CATARACT REFRACT SURG

the IOP measured 4 to 8 hours after surgery (odds ra-tio, 9.26; 95% confidence interval, 1.98–43.36; P Z .005).The corneal endothelial cell loss of 20% correspondedto the 70th centile of the percentage cell loss distribu-tion in the entire sample.

DISCUSSION

We found a corneal endothelial cell loss of 14.5% 3months after phacoemulsification in eyes with occlud-able angles. A shorter AL and higher IOP 4 to 8 hoursafter surgery were the best set of independent predic-tors of greater loss of corneal endothelial cells. The lossby 14.5% is greater than the average 4% to 10% reduc-tion in corneal endothelial cell density after phacoe-mulsification in nonglaucomatous eyes,14,15 althoughit approximates the loss of 18.3% in 1 study of 18postiridotomy eyes with PACG in which phacoemul-sification was performed for cataract or uncontrolledIOP.6

Compared with age-matched nonglaucomatouseyes, eyes with PAC or PACG may have a lower cor-neal endothelial cell density associated with a historyof acute attack of angle closure and the mean IOP inthe preceding 24 months.8,9 In the present study, wefound no difference in the preoperative corneal endo-thelial cell indices between the PACG, PAC, and PACSgroups or between eyes with and eyes without a his-tory of acute angle closure. Possible explanations arethat the preoperative IOPs were comparable betweenthe 3 diagnostic groups and the acute angle closure,if it occurred, happened at least 1 year earlier.

Postoperative IOP spikes after uneventful phaco-emulsification are common. Ahmed et al.16 reportIOP elevations to 28 mm Hg or higher in 73 (18.4%)nonglaucomatous eyes and 32 (46.4%) glaucomatouseyes 3 to 7 hours after phacoemulsification; 14 (3.6%of total) nonglaucomatous eyes and 13 (18.8% of total)

Table 1. Preoperative and postoperative clinical data.

Mean G SD

Parameter Preoperative 3 Mo Postoperative P Value

Visual acuity (logMAR) 0.39 G 0.20 0.10 G 0.12 !.001Intraocular pressure (mm Hg) 14.29 G 3.52 11.81 G 3.01 !.001Glaucoma medication (n) 1.35 G 1.22 0.44 G 0.77 !.001*Corneal endothelium

Cell count (cells/mm2) 2253.1 G 521.8 1903.9 G 658.3 !.001Cell area (100 mm2) 475.2 G 146.0 697.1 G 311.8 !.001Coefficient variation of cell area 33.6 G 8.9 35.6 G 8.9 .19Hexagonal cells (%) 49.8 G 14.8 50.4 G 17.8 .89

*Wilcoxon signed rank test.

- VOL 34, JANUARY 2008

49CORNEAL ENDOTHELIUM AFTER PHACOEMULSIFICATION IN EYES WITH OCCLUDABLE ANGLES

Table 2. Correlations between percentage of corneal endothelial cell loss and preoperative variables as well as early postoperative IOP.

Parameter Mean G SD Correlation Coefficient P Value

Preoperative variableAge (years) 73.9 G 7.0 0.199 .15IOP (mm Hg) 14.29 G 3.52 �0.266 .05Glaucoma medication (n) 1.35 G 1.22 �0.184 .18*

Visual field defect (PSD) (dB) 5.13 G 3.53 �0.010 .94Angle width† 0.42 G 0.49 �0.099 .57*

Radius of corneal curvature (mm) 7.61 G 0.31 �0.349 .03Actual ACD (mm) 1.78 G 0.25 0.106 .51Nuclear opalescence (LOCS III) 4.77 G 0.47 0.324 .04*

Lens thickness (mm) 5.09 G 0.39 �0.253 .11Axial length (mm) 22.62 G 0.94 �0.410 .008Relative lens positionz 2.14 G 0.11 0.216 .17Lens thickness/axial length ratio � 10 2.24 G 0.19 �0.040 .80Central corneal thickness (mm) 545.4 G 34.6 0.094 .99Corneal endothelium

Cell count (cells/mm2) 2253.1 G 521.8 0.075 .59Cell area (100 mm2) 475.2 G 146.0 0.176 .29Coefficient variation of cell area 33.6 G 8.9 0.085 .61Percentage of hexagonal cell 49.8 G 14.8 0.010 .95

Early postoperative IOP (mm Hg)At 4–8 hours (range) 17.68 G 5.93 (7–31) 0.322 .04The following morning (range) 14.51 G 4.50 (7–30) 0.474 .002

ACD Z anterior chamber depth; IOP Z intraocular pressure; LOCS III Z Lens Opacities Classification System III; PSD Z pattern standard deviation*Spearman rank correlation†Mean angle width was calculated by adding the grade in each quadrant and dividing the sum by 4. The angle was graded according to the Shaffer-Etienneclassification.13zRelative lens position Z (actual anterior chamber depth C 1/2 lens thickness)/axial length

glaucomatous eyes had an IOP above 40 mm Hg.In our study, the IOP varied between 7 mm Hg and31 mm Hg at 4 to 8 hours and the morning after sur-gery, with 15% and 5% of patients, respectively, hav-ing an IOP higher than 22 mm Hg. As low as theincidence and magnitude of the pressure spikes mayhave been, the detrimental effect of postoperativehigh IOP on the corneal endothelium was evident.Similarly, in a series of 21 eyeswith PACGhaving pha-coemulsification for vision-disturbing cataract, Laiet al.7 found that the 2 eyes developing corneal decom-pensation were among the 3 eyes that had IOP spikes(32 to 35 mm Hg) on day 1, although the spikes werecontrolled within 24 hours.

Phacoemulsification is associated with free radicalformation in the anterior chamber,17,18 which maylead to postoperative corneal endothelial cell loss byinducing apoptosis and necrosis of the endothelium.19

The oxidative stress achieves the highest level in thefirst postoperative 24 hours and then tapers gradu-ally.20 Meanwhile, elevated IOP is associated with in-creased oxidative stress in the anterior chamber ofhuman eyes.21 Accordingly, early postoperative IOPspikes may aggravate the oxidative damage to corneal

J CATARACT REFRACT SURG

endothelial cells in eyes with exhausted antioxidantmechanisms in the anterior segment. This may explainwhy in our study, early postoperative IOP was identi-fied as a predictor of greater postoperative cornealendothelial cell loss after adjusting for the probableconfounding effect of phacoemulsification time andnuclear opalescence.

Table 3. Final multiple linear regression model* for predictingpostoperative corneal endothelial cell loss.

Variable Coefficient (95% CI) P Value

Axial length (mm) �0.080 (�0.129 to �0.032) .002IOP 4–8 hourspostop (mm Hg)

0.014 (0.006 to 0.022) .001

Nuclear opalescence(LOCS III)

0.112 (0.000 to 0.224) .05

Phacoemulsificationtime (seconds)

0.000 (�0.002 to 0.001) .692

CI Z confidence interval; IOP Z intraocular pressure; LOCS III Z LensOpacities Classification System III*Model R2 Z 40%, P Z .001

- VOL 34, JANUARY 2008

50 CORNEAL ENDOTHELIUM AFTER PHACOEMULSIFICATION IN EYES WITH OCCLUDABLE ANGLES

Mechanical contact with nuclear fragments has beenproposed as the major cause of endothelial injury dur-ing phacoemulsification.14,22,23 Eyes with a narrowerworking space for phacoemulsification, such as thosewith a shallower ACD and shorter AL, are at a greaterrisk for themechanical contact and thusmay losemorecorneal endothelial cells after surgery. However, onlyAL was found to be associated with the postoperativecorneal endothelial cell loss in the present study. Onepossible reason we did not find a correlation betweenthe preoperative ACD and postoperative corneal en-dothelial cell loss is that the actual ACD distributionin our patients was within a narrow range, from 1.35to 2.49 mm. In addition, the accuracy of A-scan ultra-sound biometry is approximately 0.1 mm and is lim-ited further in eyes with reduced anterior segmentdimensions.24,25 Walkow et al.26 also found that AL,but not ACD, was an independent predictor of cornealendothelial cell loss 12 months after phacoemulsifica-tion in nonglaucomatous eyes.

Eyes with a shallower actual ACD preoperativelyhad lower IOP 4 to 8 hours postoperatively and atthe 3-month follow-up visit (P Z .008 and P Z .01, re-spectively; Pearson correlation). In nonglaucomatouseyes, Issa et al.27 also found that the reduction in IOPafter phacoemulsification was inversely related to thepreoperative ACD, although themechanism remaineduncertain. It is likely that the detrimental effect of pos-sible mechanical contact with the nuclear fragmentswas counterbalanced by the protective effect of lowerpostoperative IOP on the corneal endothelium in eyeswith a shallow ACD. Thus, a correlation between pre-operative ACD and corneal endothelial cell loss 3months after surgery could not be identified.

The study was limited by the small sample size,short postoperative follow-up, and single ethnic back-ground of all patients and because the function of thecorneal endothelial cells was not evaluated. In addi-tion, only 16.4% of the patients had a history of acuteattack of angle closure that occurred at least 1 year be-fore phacoemulsification; thus, our findings cannot beextrapolated to those with recent acute attacks.

In conclusion, preoperative AL measurement wasa useful predictor of corneal endothelial cell loss 3months after phacoemulsification in eyes with occlud-able angles. Extreme caution should be exercised dur-ing phacoemulsification in eyes with an AL less than22.6 mm. Close monitoring and strict control of post-operative IOP within the first 24 hours may be benefi-cial to prevent the newly injured corneal endothelialcells from additional trauma. To our knowledge, thisstudy is the first to focus on eyes with occludable an-gles. Further studies of a larger population of differentethnicity and with a longer follow-up are needed toconfirm the role of AL and early postoperative IOP

J CATARACT REFRACT SUR

in predicting corneal endothelial cell loss after phaco-emulsification in eyes with occludable angles.

REFERENCES1. Foster PJ, Johnson GJ. Glaucoma in China: how big is the

problem? Br J Ophthalmol 2001; 85:1277–1282

2. Marchini G, Pagliarusco A, Toscano A, et al. Ultrasound biomi-

croscopic and conventional ultrasonographic study of ocular

dimensions in primary angle-closure glaucoma. Ophthalmology

1998; 105:2091–2098

3. Nolan WP, Foster PJ, Devereux JG, et al. YAG laser iridotomy

treatment for primary angle closure in east Asian eyes. Br J Oph-

thalmol 2000; 84:1255–1259

4. Nonaka A, Kondo T, Kikuchi M, et al. Cataract surgery for resid-

ual angle closure after peripheral laser iridotomy. Ophthalmol-

ogy 2005; 112:974–979

5. Gunning FP, Greve EL. Uncontrolled primary angle closure glau-

coma: results of early intercapsular cataract extraction and

posterior chamber lens implantation. Int Ophthalmol 1991; 15:

237–247

6. Kubota T, Toguri I, Onizuka N, Matsuura T. Phacoemulsification

and intraocular lens implantation for angle closure glaucoma

after the relief of pupillary block. Ophthalmologica 2003; 217:

325–328

7. Lai JSM, Tham CCY, Chan JCH, et al. The clinical outcomes of

cataract extraction by phacoemulsification in eyes with primary

angle-closure glaucoma (PACG) and co-existing cataract; a pro-

spective case series. J Glaucoma 2006; 15:47–52

8. Sihota R, Lakshmaiah NC, Titiyal JS, et al. Corneal endothelial

status in the subtypes of primary angle closure glaucoma. Clin

Exp Ophthalmol 2003; 31:492–495

9. Gagnon M-M, Boisjoly HM, Brunette I, et al. Corneal endothelial

cell density in glaucoma. Cornea 1997; 16:314–318

10. Foster PJ, Buhrmann R, Quigley HA, Johnson GJ. The definition

and classification of glaucoma in prevalence surveys. Br J Oph-

thalmol 2002; 86:238–242

11. Liu CJ-L, Cheng C-Y, Wu C-W, et al. Factors predicting intraoc-

ular pressure control after phacoemulsification in angle-closure

glaucoma. Arch Ophthalmol 2006; 124:1390–1394

12. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities

Classification System III; the Longitudinal Study of Cataract

Study Group. Arch Ophthalmol 1993; 111:831–836

13. European Glaucoma Society. Terminology and Guidelines for

Glaucoma, 2nd ed. Savona, Italy, Editrice DOGMA, 2003;

Intro1–Intro13

14. Bourne RRA, Minassian DC, Dart JKG, et al. Effect of cataract

surgery on the corneal endothelium; modern phacoemulsification

compared with extracapsular cataract surgery. Ophthalmology

2004; 111:679–685

15. Zetterstrom C, Laurell C-G. Comparison of endothelial cell loss

and phacoemulsification energy during endocapsular phacoe-

mulsification surgery. J Cataract Refract Surg 1995; 21:55–58

16. Ahmed K II, Kranemann C, Chipman M, Malam F. Revisiting

early postoperative follow-up after phacoemulsification. J Cata-

ract Refract Surg 2002; 28:100–108

17. Cameron MD, Poyer JF, Aust SD. Identification of free radicals

produced during phacoemulsification. J Cataract Refract Surg

2001; 27:463–470

18. Takahashi H, Sakamoto A, Takahashi R, et al. Free radicals in

phacoemulsification and aspiration procedures. Arch Ophthal-

mol 2002; 120:1348–1352

19. Hull DS. Oxygen free radicals and corneal endothelium. Trans

Am Ophthalmol Soc 1990; 88:463–511; Available at:. http://

www.pubmedcentral.nih.gov/picrender.fcgi?artidZ1298599&

blobtypeZpdf. Accessed October 8, 2007

G - VOL 34, JANUARY 2008

51CORNEAL ENDOTHELIUM AFTER PHACOEMULSIFICATION IN EYES WITH OCCLUDABLE ANGLES

20. De Biaggi CP, Barros PSM, Silva VV, et al. Ascorbic acid levels

of aqueous humor of dogs after experimental phacoemulsifica-

tion. Vet Ophthalmol 2006; 9:299–302

21. Ferreira SM, Lerner SF, Brunzini R, et al. Oxidative stress

markers in aqueous humor of glaucoma patients. Am J Ophthal-

mol 2004; 137:62–69

22. Hayashi K, Hayashi H, Nakao F, Hayashi F. Risk factors for cor-

neal endothelial injury during phacoemulsification. J Cataract

Refract Surg 1996; 22:1079–1084

23. Beesley RD, Olson RJ, Brady SE. The effects of prolonged pha-

coemulsification time on the corneal endothelium. Ann Ophthal-

mol 1986; 18:216–219; 222

24. Binkhorst RD. The accuracy of ultrasonic measurement of the

axial length of the eye. Ophthalmic Surg 1981; 12:363–365

25. Giers U, Epple C. Comparison of A-scan device accuracy. J Cat-

aract Refract Surg 1990; 16:235–242

J CATARACT REFRACT SUR

26. Walkow T, Anders N, Klebe S. Endothelial cell loss after phacoe-

mulsification: relation to preoperative and intraoperative param-

eters. J Cataract Refract Surg 2000; 26:727–732

27. Issa SA, Pacheco J, Mahmood U, et al. A novel index for predict-

ing intraocular pressure reduction following cataract surgery.

Br J Ophthalmol 2005; 89:543–546

First author:Yu-Chieh Ko, MD

Department of Ophthalmology,Taipei Veterans General Hospital,Taipei, Taiwan

G - VOL 34, JANUARY 2008