Imaging Quiz Case 1

CLINICAL TRIALS

SECTION EDITOR: ROY W. BECK, MD, PhD

Canadian Glaucoma Study

2. Risk Factors for the Progression of Open-angle Glaucoma

Balwantray C. Chauhan, PhD; Frederick S. Mikelberg, MD; A. Gordon Balaszi, MD; Raymond P. LeBlanc, CM, MD;Mark R. Lesk, MSc, MD; Graham E. Trope, MB, PhD; for the Canadian Glaucoma Study Group

Objective: To determine systemic and ocular risk fac-tors for visual field progression in open-angle glaucoma.

Methods: In the Canadian Glaucoma Study, a multi-center prospective longitudinal study of 258 patients (131men and 127 women; median age, 65.0 years), baseline sys-temic measures included assessment of peripheral vaso-spasm and markers for hematopathology, coagulopathy,and immunopathology. Patients were followed up at4-month intervals with perimetry, optic disc imaging, anda standardized interventional protocol for intraocular pres-sure control. Univariate and proportional hazards modelswere used to identify factors that predicted progression.

Main Outcome Measure: Visual field progression withstandard automated perimetry.

Results: Median follow-up was 5.3 years, with 167 pa-tients (64.7%) completing 5 years or more and 67 pa-tients (26.0%) completing 7 years or more. Abnormalbaseline anticardiolipin antibody levels (hazard ratio [HR],

3.86; 95% confidence interval [CI], 1.60-9.31), higherbaseline age (HR per year, 1.04; 95% CI, 1.01-1.07), fe-male sex (HR, 1.94; 95% CI, 1.09-3.46), and higher meanfollow-up intraocular pressure (HR per 1 mm Hg, 1.19;95% CI, 1.05-1.36) before progression were associatedwith progression.

Conclusions: The Canadian Glaucoma Study identi-fied 4 independent predictive factors for glaucomatousfield progression.

Application to Clinical Practice: While confirmingthe importance of intraocular pressure in glaucoma pro-gression, this study determined other risk factors that meritawareness.

Trial Registration: clinicaltrials.gov Identifier:NCT00262626

Arch Ophthalmol. 2008;126(8):1030-1036

C LINICAL TRIALS HAVE CON-firmed the importance ofintraocular pressure (IOP)in the development1 ofopen-angle glaucoma and

its progression.2 Furthermore, treatmentto reduce IOP was shown to reduce the riskof development of glaucoma from 9.5% to4.4% in the Ocular Hypertension Treat-ment Study,3 and to reduce progression ofclinically manifest glaucoma from 27% to

12% in the Collaborative Normal Ten-sion Glaucoma Study (CNTGS)4 and from62% to 45% in the Early Manifest Glau-coma Trial (EMGT).5 However, withinthe periods of observation, there werepatients whose glaucoma continued toprogress despite the prescribed therapyfor IOP lowering and those whose glau-

coma remained stable despite receiving notreatment.4,5

Clinical and scientific evidence suggeststhe existence of ocular and systemic factorsin addition to, or even independent of, IOPthatmayberesponsible for thedevelopmentand progression of glaucoma. Despite sev-eralwell-executedclinicaltrials,thereremainslittle consensus on the identity of these fac-tors.TheCNTGSshowedthatwomenwere

1.9 timesmore likely thanmentohavepro-gression,andthat self-reportedmigraine in-creasedtheprogressionriskby2.5times.TheEMGTwasunabletoconfirmtheimportanceofeithersexormigraine.Ontheotherhand,the EMGT, but not the CNTGS, found agetobeasignificant factor forprogression.TheOcularHypertensionTreatmentStudyfoundthatage,butnotsexorself-reportedmigraine,wasasignificant factor for theconversiontoglaucoma.Differences instudypopulations,

CME available online atwww.jamaarchivescme.comand questions on page 1029

See also pages 1101 and 1138

Author Affiliations are listed atthe end of this article.Group Information: Themembers of the CanadianGlaucoma Study Group arelisted on page 1036.

ARCH OPHTHALMOL / VOL 126 (NO. 8), AUG 2008 WWW.ARCHOPHTHALMOL.COM1030

©2008 American Medical Association. All rights reserved.(REPRINTED WITH CORRECTIONS)Downloaded From: http://archopht.jamanetwork.com/ by a Wegner Health Science Info Ctr & USD User on 09/11/2013

disease stage, studyprocedures, andwhetherpatientsweretreated may explain these differences; however, the iden-tification of risk factors besides IOP remains elusive.

The Canadian Glaucoma Study (CGS) is a multicenterinterventional cohort study. Its primary objective was todetermine baseline demographic and systemic risk fac-tors, including susceptibility to peripheral vasospasm, andhematologic, coagulation, and immunopathological vari-ables associated with the progression of visual field dam-age in open-angle glaucoma under an interventional pro-tocol for IOP control. The protocol was established in thismanner to minimize the influence of IOP maintenance andvariability among participants such that other risk factorscould be better elucidated. Its secondary objectives in-cluded establishing the relationship between structural andfunctional progression, determining the utility of the newertechniquesof confocal scanning laser tomographyandshort-wavelength automated perimetry, and providing practicalguidelines and paradigms for the follow-up of patients withglaucoma with respect to diagnostic tests. The last patientexamination was in 2005. The purpose of this article is todescribe the risk factors associated with glaucomatous vi-sual field progression.

METHODS

The study participants and procedures have been described else-where in detail.6 The following is a brief synopsis. The CGS is amulticenter Canadian study involving 5 hospital-based univer-sity departments and is registered with the ClinicalTrials.gov Pro-tocol Registration System (identifier NCT00262626). It was ap-proved by the research ethics committee of each participatingcenter, and informed consent was obtained from each patient. Af-ter enrollment and documentation of baseline demographic, sys-temic, and ocular measures, patients were followed up with a stan-dardized treatment protocol for IOP control and tested every4 months with functional and structural tests. Patients with vi-sual field progression were subjected to a further reduction in IOP.6

ELIGIBILITY CRITERIA

Patients were either newly or previously diagnosed as having open-angle glaucoma. Patients with pseudoexfoliation and pigmen-tary glaucoma were also enrolled if they satisfied the eligibilitycriteria. Inclusion criteria were (1) best-corrected visual acuityof 6/10 or better with the Early Treatment Diabetic RetinopathyStudy chart; (2) photographically documented glaucomatous op-tic disc changes; (3) glaucomatous visual field changes includ-ing localized visual field defects, mean deviation better than −10dB, and a positive glaucoma hemifield test; and (4) nonocclud-able anterior chamber angles. Exclusion criteria were (1) sig-nificant nonglaucomatous ocular disease; (2) chronic nonglau-comatous ocular medication; (3) systemic diseases with knowneffectsonthevisual field;(4)greater than6.00diopters(D)(equiva-lent sphere) of myopia or hyperopia, or greater than 2.50 D astig-matism; and (5) previous incisional glaucoma surgery.

DIAGNOSTIC PROCEDURESAT BASELINE AND FOLLOW-UP

There were 2 baseline visits separated by 7 to 10 days. At the firstvisit, demographic, ocular, and medical histories were recordedfollowed by a full eye examination, standard automated perim-etry (SAP) with the 30-2 full threshold program of the Hum-

phrey Field Analyzer (Carl Zeiss Meditec, Dublin, California), theshort-wavelength automated perimetry 30-2 program of the Hum-phrey Field Analyzer, and confocal scanning laser tomographywith the Heidelberg Retina Tomograph (Heidelberg Engineer-ing GmbH, Heidelberg, Germany). At the second baseline visit,SAP and short-wavelength automated perimetry were repeated.Both baseline SAP visual fields had to meet the eligibility crite-ria. Finger blood flow measurements were made with laser Dop-pler flowmetry using a protocol to determine susceptibility to pe-ripheral vasospasm.7 Patients were defined as vasospastic if theratio of maximum (after exposure to heat) to minimum (after ex-posure to cold) blood flow exceeded 7.7 Blood samples were ob-tained at baseline, and tests for clinical chemistry, general hema-tology, coagulation, and immunopathology were conducted. Afull list of the blood tests is given in the previous CGS publica-tion.6 Finally, stereo optic disc photography was performed.

Follow-up visits occurred at 4-month intervals. Eye exami-nation, SAP, short-wavelength automated perimetry, and con-focal scanning laser tomography were conducted at every visit(3 examinations annually), whereas finger blood flow mea-surements and stereo optic disc photography were conductedat 32- and 28-month intervals, respectively (3 times over 5 years).

MAINTENANCE OF IOPAND TREATMENT STEPS

Newly diagnosed patients were required to have a 30% or greaterreduction inIOP, theresultofwhichwasdefinedas the target IOP.Previouslydiagnosedpatientshadaphysician-defined target IOPbased on their history and previous rate of change of the visualfield and optic disc. If visual field progression was confirmed, pa-tients were required to have an additional 20% or greater reduc-tion fromtheprevious target IOP, the resultofwhichwas thenewtarget IOP. A stepwise treatment protocol6 from topical mono-therapy,adjunct topical therapy,argon laser trabeculoplasty, and/orsystemiccarbonicanhydrase inhibitorsandtrabeculectomywasused to achieve the target IOP. Only IOP measurements obtainedafter attaining target IOP were included in the analysis.

PROGRESSION

Glaucomatous progression was defined on the basis of visual fieldchange determined with SAP using the glaucoma change prob-ability analyses.8 Progression was suspected when 8 or more lo-cations in the total deviation change probability map, with 4 ormore clustered locations in a single hemifield, were flagged. Afirst confirmation examination was then performed to verify thechange within 7 to 10 days. Progression was confirmed when therewas an overlap of 4 or more locations, with at least 2 locationsclustered within a single hemifield. If progression was not con-firmed at this examination, a second confirmation examinationwas conducted within 7 to 10 days. Hence, 2 of 3 examinationshad to confirm visual field progression. The visual field exami-nations were reviewed by the Visual Field Reading Committee,and its decision was relayed to the referring center or physician.

STATISTICAL ANALYSES

The original sample size estimate was based on the hypothesisthat patients with peripheral vasospasm have a lower progres-sion rate compared with nonvasospastic patients.6 The esti-mated sample size of 220 patients was based on the followingassumptions: (1) a progression rate of 20% in the vasospasticgroup and 35% in the nonvasospastic group; (2) an equal num-ber of vasospastic and nonvasospastic patients; (3) a 2-sided �of .05; (4) a � of .20, and (5) a cumulative 5-year attrition rateof 35%. The analysis was based on the study eye, which was



chosen by a random selection technique at the beginning of thestudy if both eyes were eligible.

Theendpointoccurredat the firstvisual fieldprogressionand,in patients who had progression, only data up to that point wereused.Inthenewlydiagnosedpatients, IOPmeasurementsobtainedbefore the target IOP was attained were not included in the analy-ses.Toexplore thepotential effectofdifferences inexposurevari-ables leading to bias in outcomes in patients completing and notcompleting 5 years of follow-up, we compared the difference inall baseline variables in these 2 groups of patients. In addition, wecomparedIOPinthefollow-up,bothinarepeated-measuresanaly-sis of variance and at 1, 2, 3, and 4 years of follow-up in patientscompleting and not completing 5 years of follow-up.

Risk factors forprogressionwereexploredwithKaplan-Meiersurvival analyses with the log-rank test for the univariate analy-ses. We also explored whether progression rates were differentamong centers. Because IOP was the only time-dependent vari-able, it was analyzed as a covariate in the survival analysis. Themean and standard deviation of IOP in the follow-up were alsoexplored. Variables were entered into the Cox proportional haz-ards model only if P� .10 in a forward stepwise analysis and if thehazardswere judgedtobeproportionalbyexaminingthenegative-logplotsof thesurvivorfunctions.All2-wayinteractiontermswereevaluated in the model and were included if the −2� log likeli-hoodscore indicatedabettermodel fit.Hazardratioswerederivedfrom the final model, which included only the significant terms.Theseanalyseswereperformedwithacommercial softwarepack-age (SAS, version 9.0; SAS Institute Inc, Cary, North Carolina).

RESULTS

Thestudypopulationandbaselinecharacteristicshavebeendescribed in a previous publication.6 Briefly, the CGS en-rolled 258 patients (131 men and 127 women) with glau-coma.Themedian(interquartile range[IQR])ageatenroll-ment was 65.0 (55.3-72.0) years. There were 23 patients(8.9%)withpseudoexfoliationglaucomaandnopatientswithpigmentaryglaucoma.Forty-sixpatients(17.8%)werenewlydiagnosedand212(82.2%)hadbeenpreviouslydiagnosed.Atbaseline,patientshadearly tomoderatevisual fielddam-age with a median (IQR) mean deviation of the 2 baselinevisual fieldsof−4.04(−5.82to−2.40)dB.Therewerenosta-tistically significant sex differences at baseline in any ocu-lar measures studied, except that men had a higher medianuntreatedIOPcomparedwithwomen(26.0and25.0mmHg,respectively).6 Of the systemic baseline measures, womenhada3-foldhigherprevalenceof thyroiddiseaseanda2-foldhigher prevalence of migraine, while there were almost 2.5timesmorediabeticmenthanwomen.6 Theenrollmentpe-riod spanned 5 years; however, 50% and 80% of the enroll-mentoccurredwithin2.0and3.4years,respectively.Innewlydiagnosedpatients,targetIOPwasobtainedwithtopicalmedi-cations usually within 3 office visits, 1 to 3 weeks apart.

The median (IQR) follow-up was 5.3 (3.7-7.0) years,with 167 patients (64.7%) completing 5 or more years and67 (26.0%) completing 7 or more years of follow-up. Ofthe 91 patients (35.3%) not completing 5 years of follow-up, 26 (29%) were lost to follow-up, 18 (20%) withdrewbecause of poor health, 14 (15%) withdrew without stat-ing a reason, 12 (13%) found the tests too difficult or timeconsuming, 11 (12%) relocated, 5 (5%) died, 4 (4%) wereno longer able to travel to the study site, and 1 (1%) with-drew because of pregnancy. The cumulative overall pro-gression rate determined by survival analysis at 2, 3, 4, and

5 years was 11.1% (95% CI, 7.7%-16.0%), 17.9% (95% CI,13.4%-23.7%), 22.0% (95% CI, 17.0%-28.2%), and 30.7%(95%CI,24.7%-37.6%), respectively.Seventy-onepatientsreached theendpoint.Of these,visual fieldprogressionwasconfirmed in45patients (63%)at the first confirmationex-amination and 26 patients (37%) at the second confirma-tion examination. There was no difference in the survivaltimes of newly and previously diagnosed patients (P=.42).

Descriptive baseline statistics of all variables exploredin this study in the no-progression and progression groupsare shown in Table 1. Univariate survival analyses dis-closed 7 variables that were significantly associated withprogression. Of the demographic variables, only female sexand higher baseline age were associated with progression.Abnormalities in results of 2 baseline blood tests, namelyanticardiolipin antibody (ACA) and red blood cell distri-bution width, were significant. Finally, of the ocular vari-ables, higher mean deviation (better visual field) and highermean and lower standard deviation of IOP in the fol-low-up (before confirmed progression and treatment in-tervention) were related to progression. The median (IQR)standard deviation and range of IOP were 2.07 (1.5-2.8)mm Hg and 6.0 (4.0-9.0) mm Hg, respectively.

Five variables (abnormal ACA level, higher baseline age,higher mean IOP and lower standard deviation of IOP inthe follow-up, and female sex) qualified for entry in the Coxproportional hazards model. Four of these were indepen-dently predictive of visual field progression (Table2). Noneof the 2-way interaction terms was significant. Abnormalbaseline ACA level yielded a hazard ratio of 3.86, indicat-ing that patients with a positive ACA test result were al-most 4 times as likely as those with a negative result to haveprogression. While highly predictive and statistically sig-nificant (eFigure1[http://www.archophthalmol.com]),only11 (5.4%) of the 204 patients tested had an abnormal ACAlevel. Higher baseline age was also highly significant, withevery year of age adding a 4% increased independent riskof progression (Table 2).

Mean IOP in the follow-up was highly significantlyassociated with progression, with a hazard ratio of 1.19,indicating that every 1 mm Hg added a 19% increasedrisk of progression (Table 2). Most patients had a meanIOP in the follow-up of 14 to 19 mm Hg (Figure 1).The 33.3rd and 66.7th percentiles of the distributions were15.5 and 17.0 mm Hg, respectively. The survival curvesfor the whole group divided into equal thirds based onmean IOP in the follow-up are shown in Figure 2. Fi-nally, women were almost twice as likely as men to haveprogression (Table 2, Figure 3).

There were no differences between patients who did anddid not complete 5 years of follow-up in the 8 continuousbaseline variables (shown in Table 1; P� .05, group t testor Mann-Whitney test). Of the 28 categorical variablesshown in Table 1, patients who did not complete 5 yearsof follow-up had a higher proportion of individuals withabnormal random glucose level (P=.02, Fisher exact test),red blood cell count (P=.04), and hematocrit (P=.03). Noneof the other 25 variables was significantly different be-tween these groups of patients. Intraocular pressure in thefollow-up was not significantly different in patients com-pleting and not completing 5 years of follow-up in both therepeated-measures analysis (P=.91) or mean values at 1,



2, 3, and 4 years of follow-up (P� .28, group t test), sug-gesting that there was no bias with respect to IOP expo-sure in these 2 groups of patients.

Although patients with peripheral vasospasm tendedto have a lower progression rate compared with nonva-

sospastic ones, this difference failed to reach statisticalsignificance in a univariate analysis (P=.14). There weresignificantly more vasospastic women than vasospasticmen (52.1% vs 33.6%; P=.004, �2 test). The tendency of

Table 2. Hazard Ratios of Significant VariablesFrom the Cox Proportional Hazards Regression Model

Hazard Ratio(95% CI) P Value

Abnormal baseline anticardiolipinantibody

3.86 (1.60-9.31) .003

Higher baseline age, per year 1.04 (1.01-1.07) .006Higher mean IOP, per mm Hg 1.19 (1.05-1.36) .008Female sex 1.94 (1.09-3.46) .02

Abbreviations: CI, confidence interval; IOP, intraocular pressure.

0.25

0.15

0.20

0.10

0.05

0.0010 12 14 16 18 20 22

Mean IOP, mm HgPr

opor

tion

of P

atie

nts

Figure 1. Frequency distribution of mean intraocular pressure (IOP) duringfollow-up. Data obtained after confirmed progression (if applicable) wereexcluded.

1.0

0.8

0.9

0.7

0.6

0.5

0.4

0.2

0.3

0.1

0 1 2 3 4 5Follow-up, y

83 75 68 61 56 41< 15.0 mm Hg

81 69 58 51 42 37> 17.0 mm Hg85 75 67 58 54 4015.1-17.0 mm Hg

Cum

ulat

ive

Surv

ival

< 15.0 mm Hg15.1-17.0 mm Hg> 17.0 mm Hg

Figure 2. Cumulative survival of patients divided into equal thirds based on themean intraocular pressure in the follow-up. Patients with mean values in thehighest tertile had higher cumulative progression than those in the lowest tertile(P=.03, log-rank test). Of the 258 patients, 249 (96.5%) had sufficientobservations to compute a mean intraocular pressure during follow-up. Thenumber of patients at risk for progression in the 3 groups at the differentfollow-up intervals is shown below the x-axis.

Table 1. Baseline Values of Continuous and NominalVariables in No-Progression and Progression Groups

No. (%)

NoProgression

(n=187)Progression

(n=71)

Continuous variablesa

Age, y 63.2(42.8 to 78.5)

68.1(45.4 to 77.6)

Mean deviation, dB −4.3(−10.1 to −0.7)

−3.2(−8.2 to −0.5)

Untreated IOP, mm Hg 25.0(19.0 to 34.0)

25.0(19.0 to 33.4)

Baseline target IOP, mm Hg 17.0(12.2 to 22.8)

18.0(13.0 to 24.4)

Baseline FBF, arbitrary units 30.7(3.0 to 91.3)

34.2(1.7 to 87.6)

FBF ratio, warm to cold 5.2(1.6 to 25.2)

5.2(1.7 to 27.0)

Systolic blood pressure, mm Hg 130(108 to 165)

130(109 to 182)

Diastolic blood pressure, mm Hg 80 (65 to 94) 81 (62 to 96)Historyb

Female sex 89 (47.6) 38 (53.5)Family history of glaucoma 81 (43.3) 32 (45.1)Pseudoexfoliation glaucoma 13 (7.0) 10 (14.1)Diabetes mellitus 18 (9.6) 7 (9.9)Hypertension 49 (26.2) 24 (33.8)Cardiovascular disease 27 (14.4) 14 (19.7)Stroke 2 (1.1) 1 (1.4)Migraine 28 (15.0) 8 (11.3)Thyroid disease 15 (8.0) 7 (9.9)

Peripheral vasospasmb

Vasospastic response (FBF ratio �7) 78 (45.1) 24 (36.4)Vasospastic men 34 (19.7) 7 (10.6)Vasospastic women 44 (25.4) 17 (25.8)

Clinical chemistry: random glucose levelc 22 (14.2) 4 (6.8)Hematologyc

White blood cell count 14 (8.8) 5 (8.2)Red blood cell count 32 (20.5) 9 (15.0)Hemoglobin 29 (18.1) 6 (9.8)Hematocrit 36 (22.5) 10 (16.4)Mean corpuscular volume 14 (8.8) 5 (8.2)Mean corpuscular hemoglobin 31 (20.9) 15 (25.4)Mean corpuscular hemoglobin

concentration7 (4.6) 3 (5.0)

Red blood cell distribution width 29 (18.1) 6 (9.8)Platelet count 7 (4.4) 2 (3.3)Platelet volume 11 (9.3) 3 (5.6)

Coagulationc

Partial thromboplastin time 13 (8.4) 7 (11.9)Fibrinogen 33 (21.7) 15 (24.6)Euglobulin lysis time (prestress) 8 (5.5) 6 (10.7)Euglobulin lysis time (poststress) 30 (20.5) 13 (23.2)

Immunopathology c

Anticardiolipin antibody level 4 (2.7) 7 (12.1)

Abbreviations: FBF, finger blood flow; IOP, intraocular pressure.aData are given as median (5th to 95th percentile).bData are given as number (percentage). Denominators vary because of

missing data.cData are given as number abnormal (percentage). Denominators vary

because of missing data.



vasospastic patients to have a lower progression rate per-sisted when the data were stratified by sex. Because menhad a lower progression rate than women, there was asignificant difference between vasospastic men, who hadthe lowest progression rate, and nonvasospastic women,who had the highest progression rate (eFigure 2 [http://www.archophthalmol.com]) (P=.03).

COMMENT

Theprimaryobjectiveof the2major randomizedglaucomaclinical trials with nontreated arms, namely the CNTGS4

and the EMGT,5 was to determine whether IOP loweringhad a beneficial effect on the disease course. The CNTGSderived participants from clinical samples of patients withglaucoma who had statistically normal IOP, whereas theEMGTdrewnewlydiagnosedpatientspredominately fromalargepopulationscreeningprogramandexcludedpatientsonly if their mean IOP was greater than 30 mm Hg (botheyes) or greater than 35 mm Hg (at least 1 eye). Despite thedifferentstudydesignsandpopulations,bothstudiesshowedthatIOPloweringhadabeneficialeffectacrosstheIOPrangein glaucoma. Although these studies and others,9-11 eithercross-sectional or longitudinal, have been used to identifynon–IOP-dependentfactorsrelatedtoprogression,theywerenot intended to answer these questions and may have beenlimited by issues related to post hoc analyses. The CGS wasprimarilydesignedtoaddresswhether, inan interventionalprotocolthatminimizedIOPvariationamongpatients(hencetheoretically reducing the potential of IOP as a confound-ing variable), peripheral vasospasm and a variety of bloodtests thatprovidean insight into thegeneralvascularhealthofpatientswerepredictiveofvisual fieldprogressioninglau-coma. We did not analyze any data after the first end point.The effect of treatment interventions on progression willbe the subject of a future report.

The CGS identified 4 statistically significant variablesthat were independently predictive of progression, namelypositive ACA test result, higher baseline age, higher meanfollow-up IOP (before confirmed progression and treat-ment intervention), and female sex. Equally important, pa-tients with diabetes, hypertension, cardiovascular dis-ease, migraine, and pseudoexfoliation glaucoma did nothave a higher progression rate; however, the CGS had lim-ited statistical power to elucidate factors with low preva-lence. We were also not able to demonstrate that a varietyof baseline hematologic and coagulation indexes influ-enced progression in treated glaucoma.

In the CGS, the distribution of mean IOP had a rela-tively low dispersion, with the central 50% of the valuesfalling between 15 and 18 mm Hg and the central 75% be-tween 14 and 19 mm Hg. We excluded IOP measurementsonce progression had occurred, since treatment interven-tion would significantly affect the mean and standard de-viation of IOP in the follow-up. In the untreated arm of theEMGT, every 1–mm Hg decrease in IOP equated to arounda 10% decrease in risk of progression.2 Similarly, the riskof conversion to open-angle glaucoma from ocular hyper-tension was around 10% for every 1–mm Hg increase inIOP.1 Despite our best attempts to reduce the confound-ing role of IOP, mean IOP before the first end point was apowerful predictor of progression, with approximately a20% increase in risk of progression for every 1–mm Hg in-crease inmeanIOP.TheCGS is therefore inagreementwithevery major clinical trial in glaucoma over the past decadein confirming the potent effect of IOP on the progressionof glaucoma across the spectrum of IOP.3-5,12 However, asubstantial number of patients with mean follow-up IOPin the lowest tertilehadprogression, and therewasnoclear-cutIOPlevelbelowwhichprogressiondidnotoccur.Equallyimportant, while the CGS was not designed to address thisissue,anotableproportionofuntreatedpatientsdonotshowprogression by study criteria4,5 or have very slow progres-sion,13 perhaps without a significant effect on quality of life.Hence, while we and others have shown the potency of IOPin dictating glaucoma progression and the notion that ev-ery 1–mm Hg decrease translates into a significant risk re-duction, the risk reduction probably cannot be 100%. Af-ter exhausting all other options, achieving low target IOPby surgery should probably be counterbalanced with thepotential of serious surgical complications, such as hy-potony, leaky blebs, and endophthalmitis.

In the CGS, 43% of the patients were classified as vaso-spastic, while, in a previous study using the same methodsand classifications,7 65% of patients with normal-tensionglaucomaand26%ofcontrolswerecategorizedasvasospas-tic. In a cross-sectional study,9 it was postulated that vaso-spasticpatientshadamoreIOP-dependentdisease, suggest-ing that a higher IOP would exacerbate the neuropathybecause of an inability of the optic nerve head circulationto regulate blood flow in the face of lower ocular perfusionpressure. The inference from this work was that vasospas-ticpatientswouldrespondbetter toIOPreduction.TheCGSwasdesigned toaddress thishypothesis, and,on thewhole,we were unable to find convincing evidence to support it.Therewasatendencyforvasospasticpatientstohavealowerprogression rate than nonvasospastic patients; more spe-cifically, when subdivided by sex, vasospastic men had a

1.0

0.8

0.9

0.7

0.6

0.5

0.4

0.2

0.3

0.1

0 1 2 3 4 5Follow-up, y

131 115 101 95 86 68Men127 104 92 75 66 50Women

Cum

ulat

ive

Surv

ival

MenWomen

Figure 3. Cumulative survival in men and women. Although women had ahigher cumulative progression rate, the sex difference was not statisticallysignificant in a univariate analysis (P=.17, log-rank test), but was significantin the proportional hazards model. The number of men and women at risk forprogression at the different follow-up intervals is shown below the x-axis.



significantlybetteroutcomethandidnonvasospasticwom-en. However, despite the significantly higher prevalence ofvasospasm in women, the potentially protective effects ofvasospasmintreatedglaucomawere likelymitigatedby thehigher progression rate in women.

Female sex was a predictor of progression, with al-most twice as many women as men showing progression.This finding corroborates those of the CNTGS but not theEMGT. The reasons for these discrepancies are not obvi-ous, but they may have to do with genetic and environ-mental differences between the populations, as well assample selections, study procedures, and analysis meth-ods that may or may not have disclosed differential ef-fects of treatment between men and women. Postmeno-pausal women may be susceptible to hormonal factors thatpredispose them to a higher risk of progression. Evi-dence from the Rotterdam Eye Study14 and the Blue Moun-tains Eye Study15 indicate that early menopause was sig-nificantly associated with a higher prevalence of glaucoma,suggesting that endogenous estrogen may have a protec-tive effect against glaucoma in women. Aging has differ-ent effects on retrobulbar circulation in men and post-menopausal women not receiving hormone therapy,16

whereas estrogen replacement decreases ophthalmic ar-tery resistance17,18 and plasma viscosity.18 The relevanceof these findings for glaucoma progression is unclear. Iflack of estrogen were a strong factor in glaucoma, then pre-sumably the prevalence of glaucoma in postmenopausalwomen compared with age-matched men would be higher;however, population-based prevalence studies do not con-sistently support this hypothesis.19-25

Patients with an abnormal ACA level were almost 4times as likely as those with a normal ACA level to haveprogression. Although the hazard ratio for this variable wasremarkably large and highly statistically significant, onlya small minority of the patients in the CGS had a positiveACA test result and substantially more patients with a nega-tive ACA result than those with a positive one had pro-gression. Anticardiolipin is one of the antiphospholipidantibodies found to have elevated levels in patients withacquired prothrombotic syndromes.26 The ACA levels canalso be elevated in miscarriage, systemic lupus erythem-atosus, ischemic stroke, and myocardial infarction.27 Whilethere is interest in the role of ACA in rheumatology, im-munology, obstetrics, and cardiology, there is consider-able debate as to whether elevated ACA levels are the causeor the effect of a variety of clinical disorders.28 There are2 published reports on ACA levels and glaucoma; how-ever, they reported contrary findings29,30 and, because oftheir cross-sectional design, could not address the issueof causality. Since an abnormal ACA level at baseline washighly predictive of progression in the CGS, this topic mer-its further investigation, specifically in phenotyping pa-tients with positive ACA results and determining the ocu-lar and systemic factors relevant for glaucoma progression.

There are some limitations of the CGS. Thirty-five per-cent of the patients did not complete 5 years of follow-up. We were unable to find systematic evidence that thosewho completed at least 5 years of follow-up had a differ-ent exposure to potential risk factors than those who didnot. Using 2 different analyses, we determined that the IOPrisk among these 2 groups was not significantly different.

However, 3 of the 36 variables (shown in Table 1) weresignificantly different in that the group not completing5 years of follow-up contained a higher proportion of pa-tients with abnormal glucose level, red blood cell count,and hematocrit. These findings may indicate a subgroupof patients with biochemical or hematologic disorders notcompleting the follow-up. These variables were not iden-tified as significant risk factors for progression. Nonethe-less, we cannot rule out the potential of other noninves-tigated measures having a differential influence in patientscompleting and not completing the 5-year follow-up.

Progression end points based on visual field criteria havebeen subject to considerable discussion.31-34 Because of vi-sual field variability and the absence of a reference stan-dard for visual field progression, end point criteria have tobe chosen with adequate specificity to minimize false-positive end points and, at the same time, maintain goodsensitivity. Although criteria similar to those used by theCGS have good performance characteristics,33,35 we ac-knowledge the inherent limitation of perimetric end points.The CGS contained a mixture of newly and previously di-agnosed patients. In the latter group, we were unable tomonitor thenumberof treatment interventionsbefore reach-ing the target IOP in the study. All observations in the CGSoccurred after the target IOP was reached. While we can-not rule out the possibility that newly and previously di-agnosed patients may have been exposed to different risksfor progression, the progression rates in these 2 groups ofpatients were not different.

In summary, using a prospective interventional pro-tocol, the CGS identified abnormal ACA level, highermean IOP in the follow-up, higher baseline age, and fe-male sex as significant independent risk factors for vi-sual field progression in glaucoma.

Submitted for Publication: November 5, 2007; final re-vision received March 26, 2008; accepted March 30, 2008.Author Affiliations: Departments of Ophthalmology andVisual Sciences, Dalhousie University, Queen ElizabethII Health Sciences Centre, Halifax, Nova Scotia (Drs Chau-han and LeBlanc), and University of British Columbia,Vancouver General Hospital, Vancouver, British Colum-bia (Dr Mikelberg); Departments of Ophthalmology,McGill University, Royal Victoria Hospital, Montreal, Que-bec (Dr Balaszi), and Universite de Montreal (Dr Lesk);and Department of Ophthalmology and Vision Sci-ences, University of Toronto, Toronto, Ontario (Dr Trope)Canada.Correspondence: Balwantray C. Chauhan, PhD, Depart-ment of Ophthalmology and Visual Sciences, DalhousieUniversity, Second Floor, Centennial Bldg, Queen Eliza-beth II Health Sciences Centre, Halifax, NS, Canada B3H2Y9 ([email protected]).Financial Disclosure: None reported.Funding/Support: This study was supported by the E. A.Baker Foundation of the Canadian National Institutes forthe Blind (1994-2005), by the Glaucoma Research Soci-ety of Canada (2003-2005), and by unrestricted grants fromAllergan Canada (2003-2005), Merck Frosst Canada (2003-2005), and Pfizer Canada (2003-2005).Additional Information: The eFigures are available at http://www.archophthalmol.com.



REFERENCES

1. Gordon MO, Beiser JA, Brandt JD, et al. The Ocular Hypertension Treatment Study:baseline factors that predict the onset of primary open-angle glaucoma. ArchOphthalmol. 2002;120(6):714-720.

2. Leske MC, Heijl A, Hussein M, Bengtsson B, Hyman L, Komaroff E. Factors forglaucoma progression and the effect of treatment: the Early Manifest GlaucomaTrial. Arch Ophthalmol. 2003;121(1):48-56.

3. Kass MA, Heuer DK, Higginbotham EJ, et al. The Ocular Hypertension Treat-ment Study: a randomized trial determines that topical ocular hypotensive medi-cation delays or prevents the onset of primary open-angle glaucoma. ArchOphthalmol. 2002;120(6):701-713.

4. Collaborative Normal-Tension Glaucoma Study Group. Comparison of glauco-matous progression between untreated patients with normal-tension glaucomaand patients with therapeutically reduced intraocular pressures. Am J Ophthalmol.1998;126(4):487-497.

5. Heijl A, Leske MC, Bengtsson B, Hyman L, Hussein M. Reduction of intraocularpressure and glaucoma progression: results from the Early Manifest GlaucomaTrial. Arch Ophthalmol. 2002;120(10):1268-1279.

6. Canadian Glaucoma Study Group. Canadian Glaucoma Study, 1: study design, base-line characteristics, and preliminary analyses [published correction appears in CanJ Ophthalmol. 2007;42(4):647]. Can J Ophthalmol. 2006;41(5):566-575.

7. Drance SM, Douglas GR, Wijsman K, Schulzer M, Britton RJ. Response of bloodflow to warm and cold in normal and low-tension glaucoma patients. Am JOphthalmol. 1988;105(1):35-39.

8. Heijl A, Lindgren G, Lindgren A, et al. Extended empirical statistical package forevaluation of single and multiple fields: Statpac 2. In: Mills RP, Heijl A, eds. Pe-rimetry Update 1990/1. New York, NY: Kugler & Ghedini; 1991:303-315.

9. Schulzer M, Drance SM, Carter CJ, Brooks DE, Douglas GR, Lau W. Biostatisti-cal evidence for two distinct chronic open angle glaucoma populations. Br JOphthalmol. 1990;74(4):196-200.

10. Ishida K, Yamamoto T, Kitazawa Y. Clinical factors associated with progressionof normal-tension glaucoma. J Glaucoma. 1998;7(6):372-377.

11. Carter CJ, Brooks DE, Doyle DL, Drance SM. Investigations into a vascular eti-ology for low-tension glaucoma. Ophthalmology. 1990;97(1):49-55.

12. AGIS Investigators. The Advanced Glaucoma Intervention Study (AGIS), 7: therelationship between control of intraocular pressure and visual field deterioration.Am J Ophthalmol. 2000;130(4):429-440.

13. Anderson DR, Drance SM, Schulzer M. Natural history of normal-tension glaucoma.Ophthalmology. 2001;108(2):247-253.

14. Hulsman CA, Westendorp IC, Ramrattan RS, et al. Is open-angle glaucoma as-sociated with early menopause? Am J Epidemiol. 2001;154(2):138-144.

15. Lee AJ, Mitchell P, Rochtchina E, Healey PR. Female reproductive factors andopen angle glaucoma. Br J Ophthalmol. 2003;87(11):1324-1328.

16. Harris A, Harris M, Biller J, et al. Aging affects the retrobulbar circulation differ-ently in women and men. Arch Ophthalmol. 2000;118(8):1076-1080.

17. Harris-Yitzhak M, Harris A, Ben-Refael Z, Zarfati D, Garzozi HJ, Martin BJ. Estrogen-replacement therapy: effects on retrobulbar hemodynamics. Am J Ophthalmol.2000;129(5):623-628.

18. Battaglia C, Mancini F, Regnani G, Persico N, Volpe A, De Aloysio D. Hormonetherapy and ophthalmic artery blood flow changes in women with primary open-angle glaucoma. Menopause. 2004;11(1):69-77.

19. Bengtsson B. Aspects of the epidemiology of chronic glaucoma. Acta Ophthal-mol Suppl. 1981;(146):1-48.

20. Dielemans I, Vingerling JR, Wolfs RC, Hofman A, Grobbee DE, de Jong PT. Theprevalence of primary open-angle glaucoma in a population-based study in the Neth-erlands: The Rotterdam Study. Ophthalmology. 1994;101(11):1851-1855.

21. Kahn HA, Leibowitz HM, Ganley JP, et al. The Framingham Eye Study, II: asso-ciation of ophthalmic pathology with single variables previously measured in theFramingham Heart Study. Am J Epidemiol. 1977;106(1):33-41.

22. Leske MC, Connell AM, Schachat AP, Hyman L. The Barbados Eye Study: preva-lence of open angle glaucoma. Arch Ophthalmol. 1994;112(6):821-829.

23. Mitchell P, Smith W, Attebo K, Healey PR. Prevalence of open-angle glaucomain Australia. Ophthalmology. 1996;103(10):1661-1669.

24. Shiose Y, Kitazawa Y, Tsukahara S, et al. Epidemiology of glaucoma in Ja-pan—a nationwide glaucoma survey. Jpn J Ophthalmol. 1991;35(2):133-155.

25. Tielsch JM, Katz J, Sommer A, Quigley HA, Javitt JC. Family history and risk ofprimary open angle glaucoma. Arch Ophthalmol. 1994;112(1):69-73.

26. Petri M. Pathogenesis and treatment of the antiphospholipid antibody syndrome.Med Clin North Am. 1997;81(1):151-177.

27. Muir KW. Anticardiolipin antibodies and cardiovascular disease. J R Soc Med.1995;88(8):433-436.

28. Levine SR, Brey RL, Tilley BC, et al; APASS Investigators. Antiphospholipid an-tibodies and subsequent thrombo-occlusive events in patients with ischemic stroke.JAMA. 2004;291(5):576-584.

29. Kremmer S, Kreuzfelder E, Klein R, et al. Antiphosphatidylserine antibodies are el-evated in normal tension glaucoma. Clin Exp Immunol. 2001;125(2):211-215.

30. Tsakiris DA, Osusky R, Kaiser HJ, Mueri R, Flammer J, Marbet GA. Lupus anti-coagulants/anticardiolipin antibodies in patients with normal tension glaucoma.Blood Coagul Fibrinolysis. 1992;3(5):541-545.

31. Katz J, Congdon N, Friedman DS. Methodological variations in estimating ap-parent progressive visual field loss in clinical trials of glaucoma treatment. ArchOphthalmol. 1999;117(9):1137-1142.

32. LeeAC,SamplePA,BlumenthalEZ,BerryC,Zangwill L,WeinrebRN. Infrequentcon-firmation of visual field progression. Ophthalmology. 2002;109(6):1059-1065.

33. Vesti E, Johnson CA, Chauhan BC. Comparison of different methods for detectingglaucomatous visual field progression. Invest Ophthalmol Vis Sci. 2003;44(9):3873-3879.

34. Chauhan BC. Choosing endpoints in clinical studies and trials. Eye. December2007;21(suppl 1s):S34-S37. doi:10.1038/sj.eye.6702886.

35. Artes PH, Nicolela MT, LeBlanc RP, Chauhan BC. Visual field progression in glau-coma: total versus pattern deviation analyses. Invest Ophthalmol Vis Sci. 2005;46(12):4600-4606.

Canadian Glaucoma Study Group

Participating Centers

Dalhousie University/Queen Elizabeth II HealthSciences Centre, Halifax, Nova Scotia. Investigators:Balwantray C. Chauhan, PhD, Raymond P. LeBlanc, CM,MD (center principal investigator), Marcelo T. Nicolela,MD (1998-present), and Paul E. Rafuse MD, PhD. Refer-ring physicians: David M. Andrews, MD, MohammadHumayun, MD, James MacNeill, MD, Andrew C. Orr, MD,John H. Quigley, MD, and George A. Sapp, MD. Researchstaff: Christine A. MacDonald, BSc (1994-1995), Helen M.Sauveur, BSc (1995-2003), and Sara L. Lavender, MSc(2003-present).

McGill University/Royal Victoria Hospital, Mon-treal, Quebec. Investigators: A. Gordon Balazsi, MD (cen-ter principal investigator), Oscar P. Kasner, MD, and NabilE. Saheb, MD. Referring physicians: Alan J. Coffey, MD,W. Edward Connolly, MD, Marino J. Discepola, MD,Conrad C. Kavalec, MD, Susan K. Lindley, MD, and MarkMullie, MD. Research staff: Pearl Alexander, BSc (1994-2001), and Bonnie May, BSc (2001-present).

University of British Columbia/Vancouver Gen-eral Hospital. Investigators: Gordon R. Douglas, MD,Stephen M. Drance, OC, MD, and Frederick S. Mikel-berg, MD (center principal investigator). Referring phy-sicians: Jeffrey A. Blicker, MD, Robin S. Cottle, MD, andVincent Wong, MD. Research staff: Zarina Pardhan, BSc.

Universite de Montreal/Hopital Maisonneuve-Rosemont, Montreal. Investigator: Mark R. Lesk, MSc,MD (center principal investigator, 1997-present). Re-search staff: Pearl Alexander, BSc (1994-2001), and Bon-nie May (2001-2005).

University of Toronto/Toronto General Hospital,Toronto,Ontario. Investigators:YvonneM.Buys,MD, JohnG. Flanagan, PhD, and Graham E. Trope, MB, PhD (cen-ter principal investigator). Referring physicians: CatherineM. Birt, MD, Michael W. Easterbrook, MD, William G.Macrae, MD, Samuel N. Markowitz, MD, and MaureenWolpert, MD. Research staff: Emad Eskander, MD.

Resource Center

Coordinating Center, Dalhousie University/QueenElizabeth II Health Sciences Centre. Visual Field Read-ing Committee: Christine A. Macgillivray BSc (study co-ordinator), and Balwantray C. Chauhan, PhD. Data en-try clerks: Erin Hicks, Kathryn Patterson, and AndrewVassallo. Biostatistician: Pantellis Andreou, PhD.

Committees

Steering Committee: Balwantray C. Chauhan, PhD (prin-cipal investigator), Frederick S. Mikelberg, MD(co–principal investigator), A. Gordon Balazsi, MD,Raymond P. LeBlanc, CM, MD, Mark R. Lesk, MSc, MD,and Graham E. Trope, MD, PhD. Data Safety Commit-tee: Frederick S. Mikelberg, MD, Raymond P. LeBlanc,CM, MD, and Graham E. Trope, MB, PhD.



Correction for corneal polarization axis improves the discriminating power of scan-ning laser polarimetry. Am J Ophthalmol. 2002;134(1):27-33.

15. Tannenbaum DP, Hoffman D, Lemij HG, Garway-Heath DF, Greenfield DS, Cap-rioli J. Variable corneal compensation improves discrimination between normaland glaucomatous eyes with the scanning laser polarimeter. Ophthalmology. 2004;111(2):259-264.

16. Reus NJ, van Koolwijk LM, Lemij HG. Effects of inadequate anterior segment com-pensation on measurements with scanning laser polarimetry. Ophthalmic SurgLasers Imaging. 2006;37(1):54-57.

17. Weinreb RN, Bowd C, Greenfield DS, Zangwill LM. Measurement of the magni-tude and axis of corneal polarization with scanning laser polarimetry. ArchOphthalmol. 2002;120(7):901-906.

18. Choplin NT, Schallhorn SC, Sinai M, Tanzer D, Tidwell JL, Zhou Q. Retinal nervefiber layer measurements do not change after LASIK for high myopia as mea-sured by scanning laser polarimetry with custom compensation. Ophthalmology.2005;112(1):92-97.

19. Greenfield DS, Knighton RW. Stability of corneal polarization axis measure-ments for scanning laser polarimetry. Ophthalmology. 2001;108(6):1065-1069.

20. Hodapp E, Parrish RK, Anderson DR. Clinical Decision in Glaucoma. St Louis,MO: CV Mosby; 1993:52-61.

21. Weinreb RN, Shakiba S, Zangwill L. Scanning laser polarimetry to measure the

nerve fiber layer of normal and glaucomatous eyes. Am J Ophthalmol. 1995;119(5):627-636.

22. Bagga H, Greenfield DS, Feuer WJ. Quantitative assessment of atypical birefrin-gence images using scanning laser polarimetry with variable corneal compensation.Am J Ophthalmol. 2005;139(3):437-446.

23. Mai TA, Reus NJ, Lemij HG. Diagnostic accuracy of scanning laser polarimetrywith enhanced versus variable corneal compensation. Ophthalmology. 2007;114(11):1988-1993.

24. Bowd C, Medeiros FA, Weinreb RN, Zangwill LM. The effect of atypical birefrin-gence patterns on glaucoma detection using scanning laser polarimetry with vari-able corneal compensation. Invest Ophthalmol Vis Sci. 2007;48(1):223-227.

25. Bland JM. What is the origin of the formula for repeatability? http://www-users.york.ac.uk/~mb55/meas/seofsw.htm. Accessed March 15, 2007.

26. Bland JM, Altman DG. Measurement error and correlation coefficients. BMJ. 1996;313(7048):41-42.

27. Bland JM, Altman DG. Statistical methods for assessing agreement between twomethods of clinical measurement. Lancet. 1986;1(8476):307-310.

28. Medeiros FA, Doshi R, Zangwill LM, Vasile C, Weinreb RN. Long-term variabilityof GDx VCC retinal nerve fiber layer thickness measurements. J Glaucoma. 2007;16(3):277-281.

Correction

Errors in Abstract and Text. In the Clinical Trials ar-ticle titled “Canadian Glaucoma Study: 2. Risk Factorsfor the Progression of Open-angle Glaucoma,” by Chau-han et al, published in the August issue of the Archives(2008;126[8]:1030-1036), the Trial Registration iden-tifier, which was listed in 2 places, was incorrect. On page1030, the Trial Registration portion of the “Abstract”should have appeared as follows: “clinicaltrials.gov Iden-tifier: NCT00262626.” On page 1031, the third sentenceof the “Methods” section should have appeared as fol-lows: “The CGS is a multicenter Canadian study involv-ing 5 hospital-based university departments and is reg-istered with the ClinicalTrials.gov Protocol RegistrationSystem (identifier NCT00262626).” Online versions ofthis article on the Archives of Ophthalmology Web sitewere corrected on August 11, 2008.

(REPRINTED) ARCH OPHTHALMOL / VOL 126 (NO. 10), OCT 2008 WWW.ARCHOPHTHALMOL.COM1364

©2008 American Medical Association. All rights reserved.

Downloaded From: http://archopht.jamanetwork.com/ by a Wegner Health Science Info Ctr & USD User on 09/11/2013

Imaging Quiz Case 1

Documents

Transcript of Imaging Quiz Case 1