British Journal of Ophthalmology 1995; 79: 649-652

Is glaucoma associated with an increased risk ofcataract?

Esmeralda V M J Kuppens, Jaap A van Best, Caesar C Sterk

AbstractAimns-The risk of developing cataract inpatients with untreated glaucoma or withocular hypertension was evaluated bycomparing the values of lenticular auto-fluorescence and light transmission in16 patients with primary open angleglaucoma and 22 patients with ocularhypertension with those of 24 healthycontrols.Methods-Increase of lenticular auto-fluorescence and decrease oftransmissionvalues in comparison with controls wereconsidered to be precursors of cataract.The values of both variables were deter-mined by fluorophotometry. Each valuewas normalised for age by dividing it bythe value for a healthy control ofthe sameage.Results-The mean age normalised auto-fluorescence and transmission values ofall patients did not differ significantlyfrom those of the controls (difference<5°/0; p=0-6 and p=0-2, respectively). Alsothe mean age normalised autofluores-cence and transmission values betweenglaucoma and ocular hypertensionpatients did not differ significantly (p=0-8and p=0*9, respectively).Conclusion-The study indicates thatuntreated primary open angle glaucomaor untreated ocular hypertension do notseem to increase significantly the risk ofdeveloping cataract.(BrJ Ophthalmol 1995; 79: 649-652)

Department ofOphthalmology,Leiden UniversityHospital, Leiden, theNetherlandsE V M J KuppensJ A van BestC C Sterk

Correspondence to:E V M J Kuppens, MD,Department ofOphthalmology, UniversityHospital Leiden, PO Box9600, 2300 RC Leiden, theNetherlands.

Accepted for publication2 March 1995

Primary open angle glaucoma (POAG) andcataract are eye diseases which mainly occur inlater life. Both diseases are often diagnosedconcurrently in elderly patients which mightsuggest that the two diseases are more closelyrelated than just by age. Recent case controlstudiesl 2 reported that glaucoma could beconsidered as a powerful and independent riskfactor for the development of cataract. It waspointed out that glaucoma accounted for 5-6%of all cataracts. However, these studies dealtonly with glaucoma patients who had beenmedically and/or surgically treated. It is knownthat some of the topical glaucoma drugs3-6 andglaucoma filtering surgery4-9 can induce oraccelerate cataract formation. Therefore, thequestion whether glaucoma in itself can con-tribute to an increased risk of cataract remainsto be answered.'0The crystalline lens contains substances

which exhibit fluorescence. I Physiologicalage related changes in the composition andamount of these fluorophores arise in the lens

leading to an increased lenticular autofluores-cence and a decreased light transmissionvalue.'2 13 Diabetes14 15 as well as extensiveexposure to ultraviolet radiation16 are otherfactors acting upon the lens autofluorescenceand transmission. Increased lenticular auto-fluorescence and decreased transmissionvalues with respect to the values of healthycontrols are considered to be precursors ofcataract formation.14 16 17 These lenticularvariables can be measured in a few seconds byfluorophotometry with no burden to thepatient.The autofluorescence and transmission

values of the ocular lens were evaluated inuntreated POAG patients detected in differentstages of their disease in order to find outwhether glaucoma itself is associated with anincreased risk of developing cataract. Thelenses of untreated ocular hypertension(OHT) patients were also evaluated in order toanalyse the effect of chronically raised intra-ocular pressure levels on the lens.

Materials and methods

PATIENTSPatients were recruited from the outpatientdepartment of the Leiden University EyeClinic and healthy controls from coworkers ofthe clinic and their relatives.The POAG and OHT patients were selected

according to the following criteria: newlydetected untreated cases of POAG or OHT;open chamber angles; bilateral intraocularpressure (IOP) values above 21 mm Hg basedon five diurnal IOP measurements by applana-tion tonometry. POAG patients had to meetthe following criteria as well: glaucomatousvisual field defects and glaucomatous cuppingof the optic nerve. Patients and healthy con-trols who had a history of an ophthalmicdisease or who had ever used medicationknown to induce or accelerate cataract forma-tion were excluded from the study.The study was approved by the medical

ethics committee of the Leiden UniversityHospital and informed consent was obtainedfrom each individual after verbal andwritten explanation of the nature of the pro-cedure.

MEASUREMENT PROCEDURE AND ANALYSISFluorophotometric measurements werecarried out with the Fluorotron Master(Coherent Radiation Inc; Palo Alto, CA,USA) fitted with a special lens (anterior seg-ment adaptor) in order to obtain an adequate

649

Kuppens, van Best, Sterk

Ecrui

200Ca

0)

C 150C.)

0

= 1000

<:50

Vitreous J Lens AC Cornea00 2 4 6 8 10 12

Distance along the optical axis (mm)Figure 1 Fluorophotometric scan of the lenticular autofluorescence of an untreated openangle glaucoma patient. The broken vertical lines correspond with the lenticular surfaces;AC= anterior chamber; Fa= anterior autofluorescence peak; F =posterior autofluorescencepeak; Faex=extrapolated anterior autofluorescence peak; FpeX=extrapolated posterior

autofluorescence peak.

spatial resolution when measuring theanterior segment of the eye. Four fluoro-photometric scans of each crystalline lenswere performed. After scanning, the IOP was

measured in all patients by applanationtonometry. Figure 1 presents an example of a

scan, showing the fluorescence profile of thehuman crystalline lens. The lens autofluores-cence was determined from the four scans ofeach eye according to a method describedpreviously.'8 The distance measured alongthe optical axis was corrected for non-

linearity. A straight line was drawn throughthe posterior and anterior peaks of the lens(Fig 1). The intercept of this line and a

vertical line at half height of the anterior peakwas supposed to represent the actual value ofthe peak anterior autofluorescence (Faextrapolated=Faex). The intercept of thestraight line and a vertical line at half height ofthe posterior peak was supposed to representthe actual posterior peak autofluorescence (Fpextrapolated=FPex). The lens transmissionT was calculated by the equation12:T= (Fpex/Faex) 1/2.The autofluorescence and transmission

values were normalised for age dependency bytaking the ratios between the measured valuesof each participant and the calculated valuesfor healthy controls of the same age. The auto-fluorescence values for healthy controls werecalculated using a linear function of age(F=-98-9+102-age; age in years; F in ngequivalent fluorescein/ml) and the trans-mission values using an exponential functionof age (T= 1 017.(1 -e((age-ll6)/32 9)); age inyears). The variables in both functions were

Table I Data ofparticipants

No of Mean age Autofluorescence Transmission Mean IOPpatients (years (SD)) (mean (SD)) (mean (SD)) (mm Hg (SD))

POAG 16 59 (13) 1-06 (0-27) 0-96 (0 08) 27 (4)OHT 22 58 (13) 1-04 (0-25) 0-96 (0 09) 27 (5)HC 24 60 (14) 1-01 (0-25) 1-00 (0-14) -

POAG=primary open angle glaucoma; OHT=ocular hypertension; HC=healthy controls;IOP=intraocular pressure.

obtained by a weighted linear regressionanalysis and a least square approximation,respectively, to the data points of the healthycontrols.

STATISTICSThe mean age normalised autofluorescenceand transmission values of patients were com-pared with those of controls. D'Agostino's testfor departure of normality was used for theassessment of a normal distribution. Student'stwo tailed t test was used for evaluating thesignificance level.

Results

LENS AUTOFLUORESCENCEThe mean lenticular fluorescence of both eyeswas calculated for each patient and controlsince the values of the right and the lefteyes were found to be correlated significantly(linear correlation coefficients r30 9,p<0001). The data and the mean agenormalised autofluorescence values of allPOAG patients, OHT patients, and healthycontrols are presented in Table 1. Thevalues of all patients and controls weredistributed normally (p=001). The meanage normalised autofluorescence values ofPOAG and OHT patients did not differsignificantly from that of controls (p = 06 andp=07, respectively). No significant differ-ence in these autofluorescence values wasfound between the two patient groups(p=08).A weighted linear regression analysis, using

inverse squared age as weight factor, revealedthat the lenticular autofluorescence values ofPOAG, OHT patients, and controls cor-related significantly with age (r>0 9,p<0-0001). The mean increase in autofluo-rescence as a function of age ofPOAG (8-9 ngEq.ml-1*yr-1) and OHT patients (8-7 ngEq-ml- 1yr- 1) did not differ significantlyfrom the increase of controls (8-5 ngEq.ml-l yr-l; p=06 and p=09, respec-tively, see Fig 2). The mean age normalisedautofluorescence values did not correlatesignificantly with the IOP values in bothpatient groups (r<-0*1, p<0*7).

LENS TRANSMISSIONThe mean lenticular transmission value ofbotheyes was calculated since the values of the rightand the left eyes were found to be correlatedsignificantly in all patients and controls(r-ts0.7, p<0001). The mean age normalisedtransmission values of both patient groups andhealthy controls are presented in Table 1. Thevalues of patients and controls were distributednormally (p=001). The mean age normalisedtransmission values of POAG and OHTpatients did not differ significantly from that ofcontrols (p=0*3 and p=0*2, respectively, seeFig 3). Also no significant difference was foundin these values between POAG and OHTpatients (p=09).

650

Is glaucoma associated with an increased risk of cataract?

E

a)0

0

t 500

0

0c

C

0)-W

Age (years)Figure 2 Lens autofluorescence as a function of age. The solid lines were obtained byweighted linear regression analysis to the values of healthy controls. Broken lines: 95%probability intervals for healthy controls. Left panel: healthy controls; middle panel:untreated primary open angle glaucoma patients; right panel: untreated ocular hypertensionpatients.

DiscussionThe age dependency of lenticular autofluores-cence values in healthy controls, POAGpatients, and OHT patients were similar(p>05) and corresponded with those in pre-vious studies.13 18 The mean age normalisedautofluorescence and light transmission valuesof POAG and OHT patients differed less than5% from those of healthy controls (p-0 2).This implies that both untreated POAG andOHT are not associated with an increased riskof developing cataract in the axial region of thelens.14 17 19 Changes which develop outsidethis region may be overlooked in the measure-ments since fluorophotometry is based onscanning the eye in a small region along theoptical axis.The present results indicate that the IOP

levels of our POAG and OHT patients do notsignificantly influence lenticular autofluores-cence or transmission. Extremely elevated IOPlevels (for example, 50-60 mm Hg), however,like those arising in attacks of acute glaucoma,

c0

en._4E

0)-J

can induce cataract.20 The IOP levels ofour POAG and OHT patients did not exceed40 mm Hg which may explain why IOP wasnot found to be a cataract inducing factor inthis study.Our findings are supported by those of

Shaffer and coworkers4 who showed that thecataract incidence between a healthy popula-tion and a POAG population treated with pilo-carpine or carbacholine only, did not differsignificantly. They postulated that glaucoma initself did not increase the probability of devel-oping cataract. Additional evidence for ourfindings is provided by Van Buskirk2l whoreported in 1982 that of the 1000 cases oftimolol side effects registered over 3 years, only12 concerned cataract.On the other hand Harding and coworkers'

recently reported the analysis of two combinedcase-control studies: glaucoma was identifiedas a powerful and independent risk factor forcataract and was held responsible for 50/0 of allcataracts reported in England. However, thisstudy concerned medically and/or surgicallytreated glaucoma patients which may havebiased the study outcome.As pointed out in several studies,3-9 there is

evidence that both the medical and surgicaltreatment of glaucoma can cause cataract.Some of the topical antiglaucomatous medica-ments such as strong miotics like echothio-phate were held responsible for inducingcataracts in the past.3 6 The cataractous effectof frequently used miotics like pilocarpine isunclear: a previous study3 demonstrated thatpilocarpine induced formation or progressionof cataract in 10% of the POAG patientswhereas another study4 did not supplyevidence for such an effect. Surgical drainageprocedures have been recognised as a con-tributing factor to the development ofcataract'9 or merely as a cataract acceleratingfactor in patients who had already somecataract before surgery.22

It seems therefore plausible to assume thatthe treatment of glaucoma might increase theincidence of cataract in glaucomatous popula-tions. This study shows that untreated POAGand OHT do not seem to increase the risk ofdeveloping cataract.

1 Harding JJ, Egerton M, van Heyningen R, Harding RS.Diabetes, glaucoma, sex, and cataract: analysis of com-bined data from two case control studies. BrJt Ophthalmol1993; 77: 2-6.

2 Van Heyningen R, Harding JJ. A case-control study ofcataract in Oxfordshire: some risk factors. BrJ' Ophthalmol1988; 72: 804-8.

3 Axelsson U, Holmberg A. The frequency of cataract aftermiotic therapy. Acta Ophthalmol 1966; 44: 421-9.

4 Shaffer RN, Rosenthal G. Comparison of cataract incidencein normal and glaucomatous population. AmJ Ophthalmol1970; 69: 368-70.

5 Sugar HS. Postoperative cataract in successfully filteringglaucomatous eyes. Am Y Ophthalmol 1970; 69:740-6.

6 Layden WE. Cataracts and glaucoma. In: Tasman W,Jaeger EA, eds. Duane's clinical ophthalmology. Rev ed.Philadelphia: Harper & Row, 1979: 1-22.

7 Greve EL, Dake CL, Klaver JHJ, Mutsaerts EMG. Ten yearprospective follow-up of a glaucoma operation. IntOphthalmol 1985; 8: 139-46.

8 Harding JJ, Harding RS, Egerton M. Risk factors forcataract in Oxfordshire: diabetes, peripheral neuropathy,myopia, glaucoma and diarrhoea. Acta Ophthalmol 1989;67: 510-7.

9 Laatikainen L. Late results of surgery on eyes with primaryglaucoma and cataract. Acta Ophthalmol 1971; 49:281-92.

Age (years)Figure 3 Lens transmission as afunction of age. The solid lines were obtained by a leastsquare approximation procedure to the values of healthy controls. Broken lines and panelsas indicated in Figure 2.

651

Kuppens, van Best, Sterk

10 Flanagan DW. Diabetes, glaucoma, sex, and cataract.BrJr Ophthalmol 1993; 77: 1.

11 Lerman S, Borkman R. Spectroscopic evaluation andclassification of the normal, aging and cataractous lens.Ophthalmic Res 1976; 8: 335-53.

12 Van Best JA, Tjin A Tsoi EWSJ, Boot JP, Oosterhuis JA. Invivo assessment of lens transmission for blue-green lightby autofluorescence measurement. Ophthalmic Res 1985;17: 90-5.

13 Occhipinti JR, Mosier MA, Burstein NL. Autofluorescenceand light transmission in the aging crystalline lens.Ophthalmologica 1986; 192: 203-9.

14 Mosier MA, Occhipinti JR, Burstein NL. Autofluorescenceof the crystalline lens in diabetes. Arch Ophthalmol 1986;104: 1340-3.

15 Bleeker JC, van Best JA, Vrij L, van der Velde EA,Oosterhuis JA. Autofluorescence of the lens in diabeticand healthy subjects by fluorophotometry. InvestOphthalmol Vis Sci 1986; 27: 791-4.

16 van Wirdum E, Mota MC, van Best JA, Leite E, Kappelhof

JP, Faria de Abreu JR, et al. Lens transmission and auto-fluorescence in renal disease. Ophthalmic Res 1988; 20:317-26.

17 Yappert MC, Borchman D, Byrdwell WC. Comparison ofspecific blue and green fluorescence in cataractous versusnormal human lens fractions. Invest Ophthalmol Vis Sci1993; 34: 630-6.

18 Boets EPM, Kok JHC, van Best JA. The blue-green auto-fluorescence of the human ocular lens after the wear ofPMMA contact lenses. Acta Ophthalmol 1994; 72: 67-71.

19 Siik S, Airaksinen PJ, Tuulonen A, Nieminen H.Autofluorescence in cataractous human lens. InvestOphthalmol Vis Sci 1992; 33 (suppl): 1170.

20 Leydhecker W. Primare Glaukomformen. In: Glaukom: einHandbuch. 2. Auflage. Berlin: Springer-Verlag, 1973: 20.

21 Van Buskirk EM. Hazards of medical glaucoma therapy inthe cataract patient. Ophthalmology 1982; 89: 238-41.

22 Watson PG, Jakeman C, Ozturk M, Barnett MF, Barnett F,Khaw KT. The complications of trabeculectomy (a20-year follow-up). Eye 1990; 4: 425-38.

652