One Year Follow-up of Functional Recovery inNeovascular AMD During Monthly

Anti-VEGF Treatment

MARION R. MUNK, CHRISTOPHER KISS, WOLFGANG HUF, FLORIAN SULZBACHER, PHILIPP ROBERTS,TAMARA J. MITTERMULLER, STEFAN SACU, CHRISTIAN SIMADER, AND URSULA SCHMIDT-ERFURTH

� PURPOSE: To identify neurosensory recovery, testingdifferent functional variables during monthly intravitrealstandard anti–vascular endothelial growth factor (VEGF)therapy in neovascular age-related macular degeneration(AMD).� DESIGN: Prospective interventional cohort study.� METHODS: Sixty-four treatment-naıve neovascularAMD patients with subfoveal lesions were treated andexaminedmonthly fordistance visual acuity, reading acuity,maximum reading speed, and contrast sensitivity and withmicroperimetry evaluating the percentage of absolute andrelative scotoma and mean central retinal sensitivityweighted by area. Improvements in reading acuity, distanceacuity, reading speed, contrast sensitivity, mean centralretinal sensitivity, and scotoma area in dependence of age,lesion type, lesion size, and mean central retinal sensitivitywere evaluated by a random-slope and random-interceptmodel. Recovery pattern of parameters was compared bycorrelating the individual slopes of each variable.� RESULTS: Initially, a rapid short-term effect of anti-VEGF treatment was documented throughout all func-tional variables. Progressive functional gain over 1 yearwas observed for distance visual acuity (P [ .011),contrast sensitivity (P £ .0001), and mean central retinalsensitivity (P £ .0001), but not for reading acuity (P [.31) and maximum reading speed (P [ .94). Decreaseof absolute scotoma area missed statistical significanceover time (P [ .053) and also fixation stability did notimprove (P [ .08). However, lesion size influenced thecourse of absolute scotoma area (P [ .0015), whilelesion type had no effect on any visual function variableevaluated. The individual slopes of reading acuity anddistance visual acuity showed a moderate correlation;however, all other variables showed only a weak or nosignificant correlation among each other.

Supplemental Material available at AJO.com.Accepted for publication May 28, 2013.

From the Department of Ophthalmology (M.R.M., C.K., F.S., P.R.,T.J.M., S.S., C.S., U.S.E.) and Center for Medical Physics andBiomedical Engineering (W.H.), Medical University of Vienna, Vienna,Austria.

Inquiries to Christopher Kiss, Department of Ophthalmology, MedicalUniversity of Vienna, Waehringer Guertel 18-20, A-1090 Vienna,Austria; e-mail: christopher.kiss@meduniwien.ac.at

0002-9394/$36.00http://dx.doi.org/10.1016/j.ajo.2013.05.037

� 2013 BY ELSEVIER INC.

� CONCLUSION: Visual recovery in anti-VEGF therapyis reflected in a characteristic pattern of functionalchanges over time, whereas distance visual acuity doesnot seem to comprehensively reflect overall visualfunction gain. (Am J Ophthalmol 2013;156:633–643. � 2013 by Elsevier Inc. All rights reserved.)

VISUAL FUNCTION, PARTICULARLY NEAR VISUAL

function such as reading, is highly associatedwith quality of life.1–3 However, visual function

and treatment response are usually assessed in clinicalpractice and in clinical trials solely by testing distancevisual acuity (VA). Distance VA has been commonlyreported to increase during follow-up, but concomitantlypatients have reported stagnation of visual function duringtreatment.1,4 Distance VA is the method of choice tomonitor the efficacy of treatment for neovascular age-related macular degeneration (AMD), even though inactive disease impairment of reading acuity is morepronounced than distance VA and reading acuity hasbeen shown to improve more than distance VA duringtreatment.1,4,5 Reading acuity would therefore be analternative follow-up method, and the American Academyof Ophthalmology (AAO) practice pattern recommends it(http://one.aao.org/CE/PracticeGuidelines/default.aspx) asan adequate method to follow up patients with neovascularAMD. Comparison of distance VA with reading acuity,contrast sensitivity, and retinal sensitivity in studiesof different ophthalmic diseases6–8 has led to therecognition that distance VA testing is not sufficient toreflect overall visual function.9 Furthermore, the methodhas been shown to correlate poorly with other visual func-tion variables.10–12 However, most studies have been cross-sectional, examining a single time point rather than thecourse of visual function variables during treatmentfollow-up. Many studies have concentrated on the correla-tion of retinal thickness and visual function.4,13–15

We designed a prospective longitudinal study in whichpatients who had been newly diagnosed with neovascularAMD were evaluated for neurosensory function andrecovery during standard monthly anti–vascular endothe-lial growth factor (VEGF) therapy. Established classic func-tional variables were used for the evaluation.We hypothesized that measuring higher-order visual

function would improve vision assessment in patients

633ALL RIGHTS RESERVED.

with neovascular AMD and provide a realistic insightinto the benefits of anti-VEGF therapy by reflecting allof the important visual function variables that contributeto visual performance in everyday life. In addition,a more reliable investigative method should be able toexplain the mismatch between distance VA findingsand patients’ subjective visual perception, leading toimproved personalized therapeutic strategies and conse-quent enhancement of quality of life for patients withneovascular AMD.

FIGURE 1. Example of evaluation of the lesion size of an occultchoroidal neovascular ([ type I) lesion in fluorescein angiog-raphy. The contours of the leakage extent at 10 minutes weretraced with the embedded Heidelberg software and the areawas automatically calculated.

METHODS

� PATIENT SELECTION AND SETTING: In this prospectivelongitudinal study, 64 eyes of 64 consecutive subfovealtreatment-naıve patients were observed. Patients whohad recent onset of neovascular AMD and a distance VAranging from 20/40 to 20/320 and who were willing andable to attend monthly visits were enrolled. The prospec-tive study was approved by the local ethics committee,Medical University of Vienna, and registered atclinicaltrials.gov (NCT01608113). The study designadhered to the tenets of the Declaration of Helsinki andall federal laws of Austria. Patients gave written informedconsent after an extensive discussion about the purposeof the study and all the risks associated with participation.They were recruited at the Medical University of Vienna.Patients with cataract worse than grade 2 (according toLens Opacities Classification System III), amblyopia,dyslexia, Alzheimer disease, or dementia were excluded.Overall intellectual reading capabilities were tested bybinocular newspaper text reading before inclusion.

Examinations were performed before injection of anti-VEGF (baseline) and 1 day, 1 week, and 1 month afterthe initial injection. Thereafter examinations and anti-VEGF injections were given at monthly (28 days) intervals6 3 days in accordance with the study protocol. Studyexamination included testing with distance best-corrected visual acuity (BCVA), Early Treatment DiabeticRetinopathy Study (ETDRS) charts, Radner reading chartsfor reading acuity and maximum reading speed, Pelli-Robson charts for contrast sensitivity, and microperimetry.Fluorescein angiography was performed with HeidelbergRetinal Angiography (HRA; Heidelberg Engineering,Heidelberg, Germany) at baseline and after 3, 6, and12 months. Lesions detected with fluorescein angiographywere classified either as type I or type II choroidal neovas-cularizations (CNVs); classic and predominantly classicangiographic lesion types were subsumed as type II CNVand occult and minimal classic lesions as type I CNV.Lesion size (mm2) was measured at 10 minutes; for thispurpose, the contours of the leakage extent were tracedand the area was automatically calculated by the HRA-embedded software (Figure 1). Heidelberg-Spectralis

634 AMERICAN JOURNAL OF

HD-OCT (Heidelberg Engineering, Heidelberg, Germany)was performed at each monthly visit.

� STUDYVARIABLES: BCVA was tested in both eyes usingETDRS charts at a distance of 4 m. Results were recorded asthe logarithm of the minimal angle of resolution (logMAR)and visual acuity letter score.16

Radner reading chartswere used to test reading acuitywitha luminescence of 90 cd/m2 at a distance of 30 cm.The chartsoffer a reliable test for monocular reading acuity with hightest-retest reproducibility in patients with low visual func-tion. Similar to MNRead charts, each sentence is 0.1logMAR units smaller than the previous sentence and thetext is printedwithhigh contrast. In contrast to theMNReadcharts, logRAD charts consist of ‘‘sentence optotypes’’; thus,all sentences are similar in grammatical difficulty and in thenumber, length, and position of words. This reading test isavailable in several languages and was designed to correlatedistance VA and reading acuity as closely as possible.17,18

A difference of 0.1 log unit has been described betweenreading acuity measured by Radner reading charts andETDRS visual acuity in healthy eyes.19

Reading acuity was tested monocularly for both eyes.Sentences were covered and patients were asked to uncoverand read each sentence aloud, from the largest to the small-est accessible font, as quickly as possible. Reading acuitywas recorded as a logRAD score (reading equivalent oflogMAR; logRAD score ¼ logRAD þ syllablesmisread 3 0.005). Maximum reading speed was defined

OCTOBER 2013OPHTHALMOLOGY

FIGURE 2. Microperimetry grid used in this study and the location of the Early Treatment Diabetic Retinopathy Study (ETDRS)grid in respect to the stimulation loci (Left) and a representative example of a microperimetry examination of a patient with occultchoroidal neovascularization at baseline (Right). The grid consisted of 33 stimulation loci. For each microperimetry examination,the number of absolute scotoma (0 dB), severe relative scotoma (1-6 dB), mild relative scotoma (7-12 dB), and normal functionloci (13-20 dB) were counted. Following, the mean retinal sensitivity and the absolute and severe relative scotoma area was evaluatedby dedicated software. As stimulation points dependent on central or peripheral position showed different sizes in surrounding areas,mean central retinal sensitivity was weighted by area (Left). Accordingly, scotoma area evaluated by this software was also weightedaccording to respective surrounding area (Left).

as the maximum number of words read per minute (wpm),measured with a stopwatch. The best maximum readingspeed, independent of print size, was used for further anal-ysis. All eyes were refracted to achieve BCVA, includingpresbyopic correction for reading. Patients who wereunable to read the biggest letter size of 1.4 logRAD weregiven the logRAD score 2.0 with a maximum reading speedof 10 wpm.

Contrast sensitivity was evaluated using Pelli-Robsoncharts at a luminescence of 85 cd/m2 using the totalcontrast sensitivity (CS) score ([number of letters �3] 30.05, CS score).20

Microperimetry was performed with the MP-1 micro-perimeter (Nidek Advanced Vision Information System[NAVIS]; Nidek Technologies, Padua, Italy). The Carte-sian grid generated consisted of 33 test points. Thefollow-up function was used for every follow-up visit. Thefixation mark presented as a 3-degree circle and a 4-2-1staircase strategy was used. The stimulus intensity rangedfrom 0-20 dB (0 dB refers to the strongest signal intensityof 127 cd/m2) in 1-dB steps.

Stimulation loci representing an absolute scotoma(0 dB) were counted at every visit. Loci with sensitivityfrom 1-6 dB were considered severe relative scotoma andthose with sensitivity from 7-12 dB as mild relativescotoma. Values above 13 dB were considered normal

VOL. 156, NO. 4 FUNCTIONAL RECOVERY IN WET AGE-R

(¼ normal function loci). Afterwards, microperimetrydata were exported in mpd format and analyzed usingproprietary software developed by one of the authors(C.S.) to automatically transfer retinal sensitivity valuesfor evaluation of mean retinal sensitivity according to the1-, 3-, and 6-mm ETDRS grids.21,22 Thus, stimulationloci within a radius of 0.5 mm were automaticallycounted for the 1-mm central ETDRS grid, loci between0.5 and 1.5 mm were automatically counted for the3-mm ETDRS grid, and stimuli between 1.5 and 3 mmwere automatically counted for the 6-mm ETDRS grid bythe respective software (Figure 2). The software then notonly summarized mean central retinal sensitivity valuesbut also calculated the mean retinal sensitivity value byweighting the sensitivity data by area. Besides mean retinalsensitivity, absolute and severe relative scotoma size wasalso evaluated by this software through interpolation.Absolute scotoma size was evaluated by multiplying eachstimulation locus <_0 with its associated area. Correspond-ingly, each stimulation locus <_6 dB was multiplied withits associated area for the evaluation of the size of severerelative scotoma. The associated area varied according tothe location of the loci, as central loci are closer to eachother than peripheral loci (Figure 2). Thus, respectivescotoma size was automatically calculated in regard to thetotal area tested.

635ELATED MACULAR DEGENERATION

TABLE 1.Mean6 SD of Evaluated Visual Function Variablesat Each Visit in Patients With Neovascular Age-Related

Macular Degeneration Treated With Monthly Standard Anti–

Vascular Endothelial Growth Factor Treatment

Reading Acuity

(logRAD Score)

Distance VA

(logMAR)

Maximum

Reading

Speed (wpm)

Contrast

Sensitivity

(CS Score)

Baseline 0.93 6 0.6 0.61 6 0.32 108 6 63 1.06 6 0.22

1 day 0.85 6 0.53 0.55 6 0.25 120 6 62 1.06 6 0.21

1 week 0.85 6 0.59 0.51 6 0.26 119 6 66 1.08 6 0.26

1 month 0.82 6 0.57 0.50 6 0.29 118 6 60 1.13 6 0.22

2 months 0.70 6 0.49 0.45 6 0.30 125 6 61 1.17 6 0.20

3 months 0.72 6 0.53 0.48 6 0.31 129 6 64 1.20 6 0.21

4 months 0.68 6 0.49 0.44 6 0.29 124 6 55 1.25 6 0.20

5 months 0.70 6 0.57 0.43 6 0.32 121 6 61 1.22 6 0.19

6 months 0.67 6 0.52 0.44 6 0.31 124 6 58 1.23 6 0.22

7 months 0.69 6 0.55 0.43 6 0.34 122 6 64 1.23 6 0.23

8 months 0.71 6 0.54 0.42 6 0.28 119 6 68 1.25 6 0.20

9 months 0.69 6 0.54 0.43 6 0.33 126 6 63 1.24 6 0.24

10 months 0.81 6 0.62 0.44 6 0.35 116 6 70 1.25 6 0.20

11 months 0.66 6 0.52 0.42 6 0.32 125 6 65 1.24 6 0.27

12 months 0.80 6 0.61 0.46 6 0.36 111 6 66 1.22 6 0.22

logMAR¼ logarithm of minimal angle of resolution; VA¼ visual

acuity; wpm ¼ words per minute.

Fixation stability within 2 degrees and 4 degrees accord-ing to the MP-1 microperimetry examination set-ups wasrecorded at baseline and 3, 6, and 12 months after theinitial treatment. The MP-1 automatically acquires about25 fixation points per second. In accordance with a studyby Fujii and associates, a finding of more than 75% of thefixation points inside the 2-degree-diameter circle wasclassified as a ‘‘stable’’ fixation, more than 75% fixationpoints inside the 4-degree-diameter circle and less than75% inside the 2-degree-diameter circle as a ‘‘relativelyunstable’’ fixation, and less than 75% inside the 4-degree-diameter circle as an ‘‘unstable’’ fixation.23,24

� STATISTICAL ANALYSES: Distance VA (logMAR),contrast sensitivity (CS score), reading acuity (logRADscore), maximum reading speed (maximum wpm), andmean retinal sensitivity were the main outcome variables.

Time, age, and lesion type (type I vs type II) were inde-pendent factors for the respective functional variables;intercept and slope of a linear trend were random factors.In the prespecified analysis, data were analyzed by a randomslope and random intercept model with an unstructuredcovariance matrix.25 The difference in fixation stability(stable, relatively unstable, unstable) between baselineand after 1 year was analyzed with a Wilcoxon signedrank test. Further, to correlate lesion size, lesion type, andabsolute scotoma size, the effect of lesion size and lesiontype adjusted for age and time on absolute scotoma sizewas analyzed using a similar random-slope–random-intercept model to that mentioned above, with the mainoutcome variable ‘‘absolute scotoma size’’ as a dependentvariable and lesion type (type I vs type II), lesion size,time, and age as independent variables.

Pearson correlations were used to correlate the timecourse of the different functional variables with the indi-vidual slopes of the variables.

Statistical analyses were conducted with the statisticsprogram R (www.r-project.org). Figures were also drawnwith R. Descriptive statistics are reported as mean 6 stan-dard deviation for continuous data and as absolutefrequency (percentage) for categorical data. The level ofsignificance was set at P <_ .05.

RESULTS

� BASELINE CHARACTERISTICS: Sixty-four eyes of 64consecutive patients were included in this prospectivelongitudinal study. All patients received monthly treat-ments following a fixed regime. There were 3 early discon-tinuations: 2 patients left the study after 4 months(1 because of prostate carcinoma and 1 after complaintsabout the time-consuming study procedure), and 1 patientleft the study after 6 months after a myocardial infarction.Because of missing data points, overall 61 patients wereincluded in the per-protocol analysis. Mean age was

636 AMERICAN JOURNAL OF

77.26 8.5 years. Thirty-three patients presented an occultangiographic, 12 a classic angiographic, 12 a minimallyclassic, and 4 a predominantly classic lesion type. Thus,45 patients showed type I lesions and 16 type II lesions.The mean lesion size at baseline was 7.5 6 6.1 mm2:7.9 6 6.1 mm2 in occult, 6 6 5.6 mm2 in classic, 7.1 66.7 mm2 in minimally classic, and 10.8 6 6.6 mm2 inpredominantly classic lesions. Lesion size did not differbetween lesion types (2-sample t test: type I: 7.7 6 6.2,type II: 7.1 6 6.0, P ¼ .72).

� COURSE OF VISUAL FUNCTION VARIABLES DURINGOBSERVATIONAL PERIOD: The mean values of the vari-ables for functional testing evaluated are shown inTables 1 and 2. In general, all functional variablesresponded rapidly to treatment (Tables 1 and 2, Figures 3and 4). However, progressive functional gain over 1 yearwas only seen in distance VA (P ¼ .011), contrast sensi-tivity (P <_ .0001), and mean central retinal sensitivity(P <_ .0001). No gain in reading acuity (P ¼ .31) andmaximum reading speed (P ¼ .91) was seen. CNV lesiontype had no influence on the course of any visual functionparameter, whereas age affected contrast sensitivity(P¼ .046) and central retinal sensitivity (P¼ .02). Furtherdetails are provided in Tables 3 and 4.

� CHANGE OF FIXATION STABILITY: At baseline, 62.7%(37 of 59) of the eyes presented a relatively unstable,27.1% (16 of 59) an unstable, and 10.2% (6 of 59) a stablefixation. One year after continuous treatment, 72.3%

OCTOBER 2013OPHTHALMOLOGY

TABLE 2. Mean 6 SD of Central Retinal Sensitivity Parameters at Each Visit in Patients With Neovascular Age-Related MacularDegeneration Treated With Monthly Standard Anti–Vascular Endothelial Growth Factor Treatment

Absolute Scotoma

Sizea (mm2)

Severe Relative

Scotoma Sizeb (mm2) Absolute Scotomac (n)

Severe Relative

Scotomad (n)

Mild Relative

Scotomae (n) Normal Function Locif (n) Mean CRS (dB)

Baseline 4.7 6 4.9 8.9 6 5.4 10.5 6 10.0 8.5 6 5.9 9.1 6 6.5 5.0 6 6.8 7.3 6 4.5

1 day 5.1 6 5.5 9.1 6 5.7 9.3 6 10.6 8.7 6 7.6 9.0 6 6.8 5.5 6 8.1 6.9 6 4.6

1 week 3.6 6 4.1 7.6 6 5.3 8.5 6 9.8 8.9 6 6.4 9.6 6 6.4 5.7 6 7.6 7.8 6 4.5

1 month 0.82 6 0.57 0.50 6 0.29 7.4 6 8.4 8.0 6 6.0 9.0 6 5.3 8.0 6 8.9 8.8 6 4.7

2 months 3.4 6 3.8 6.4 6 5.2 6.8 6 7.6 6.2 6 5.1 9.7 6 6.0 9.9 6 9.5 9.6 6 4.7

3 months 2 6 2.5 5.1 6 4.6 5.0 6 6.1 6.6 6 5.9 11.3 6 7.1 9.9 6 9.7 10.2 6 4.3

4 months 2.4 6 3.4 4.5 6 4.7 5.2 6 7.2 4.8 6 4.8 9.9 6 5.8 13.1 6 10.4 11.2 6 4.2

5 months 2.2 6 2.5 4.7 6 4.3 5.1 6 6.3 4.3 6 4.4 10.5 6 6.3 13.1 6 10 10.8 6 4.4

6 months 2.6 6 3.7 4.8 6 5 6.0 6 8.2 4.3 6 4.7 10.1 6 6.8 12.6 6 10.1 10.8 6 4.3

7 months 2.6 6 3.4 5.4 6 5.8 6.4 6 7. 7 5.7 6 5.8 9.7 6 6.2 11.5 6 10.2 10.5 6 4.9

8 months 2.4 6 3.2 4.8 6 4.8 5.1 6 7.3 4.5 6 5.4 9.7 6 6.6 13.7 6 10.9 11.0 6 4.5

9 months 3 6 3.7 5.4 6 5.3 5.9 6 7.8 4.67 6 4.9 9.7 6 6.5 12.7 6 10.5 10.6 6 4.7

10 months 2.7 6 3.2 5 6 4.5 6.1 6 7.6 4.4 6 4.3 8.9 6 5.8 13.6 6 11.1 10.8 6 4.8

11 months 7,9 3.7 6 3.8 4.8 6 7.3 4.0 6 4.3 9.9 6 7.2 14.3 6 11.0 11.6 6 4.4

12 months 2.4 6 3.2 4.7 6 4.5 5.8 6 7.9 4.3 6 3.8 9.9 6 6.8 13.1 6 10.8 10.8 6 4.5

CRS ¼ central retinal sensitivity.aMean corresponding area of microperimetry test points presenting 0 dB threshold value at respective visit.bMean corresponding area of microperimetry test points presenting a 0-6 dB threshold value.cMean number of microperimetry test points (max. number: 33) presenting 0 dB threshold value at respective visit.dMean number of microperimetry test-points (max. number: 33) presenting a 1-6 dB threshold value.eMean number of microperimetry test points presenting a 7-12 dB threshold value.fMean number of microperimetry test points presenting a 13-20 dB threshold value.

(34 of 47) exhibited a relatively unstable, 12.8% (6 of 47)an unstable, and 14.9% (7 of 47) a stable fixation(Wilcoxon signed rank test: P ¼ .08).

� COURSE OF ABSOLUTE SCOTOMA SIZE AND CORRELA-TION TO LESION SIZE AND LESION TYPE: In the statisticalmixed model applied herein, decrease of absolute scotomasize missed statistical significance over time and absolutescotoma did not significantly diminish in size undermonthly treatment (P ¼ .053). However, the initialCNV lesion size, but not the CNV lesion type, influencedthe course of absolute scotoma area: the larger the CNVlesion at baseline, the slower and smaller the reduction ofscotoma area under treatment (P ¼ .0015) (Table 5, andSupplemental Figure, available at AJO.com).

� CORRELATIONS OF VISUAL FUNCTION VARIABLES:

The correlations between the individual slopes of readingacuity, distance VA, maximum reading speed, contrastsensitivity, and mean central retinal sensitivity arepresented in Table 6. The individual slopes of readingacuity and distance VA as well as the slopes of readingacuity and maximum reading speed were moderately corre-lated (r ¼ 0.62, P <_ .0001 and r ¼�0.681, P <_ .0001) (seealso Figure 5). All the other slopes of the functional vari-ables correlated weakly (r ¼ 0.3-0.5) or did not correlate(r < 0.3) with each other.

VOL. 156, NO. 4 FUNCTIONAL RECOVERY IN WET AGE-R

DISCUSSION

IN AMD PATIENTS, DISTANCE VA ALONE DOES NOT SEEM TO

reflect the improvement in the different vision-relatedtasks of everyday life. Contrast sensitivity and readingacuity as well as retinal sensitivity have been suggested tobe more sensitive tests for disease activity and quality oflife.4,6,7,26–28 Therefore, this study was performed todescribe the course of several additional visual functionparameters besides distance VA in neovascular AMDpatients under anti-VEGF treatment.In this study, distance VA, contrast sensitivity, and

mean central retinal sensitivity improved progressivelyover 12 months of monthly treatment, whereas readingacuity and maximum reading speed did not improve. Wefound a similar mean increase in distance VA to thatreported in recent large multicenter trials,29–32 suggestingthat our study population presented exemplary anti-VEGF outcomes under optimized conditions. Conse-quently, the improvements in the other visual functionvariables should also be representative.Contrast sensitivity and retinal sensitivity improved

concordantly to distance VA in this study. Patients inour study received monthly injections and gained 6.4contrast-sensitivity letters, equivalent to a mean of 0.17log contrast sensitivity. The 64 patients who receiveda mean of 7 bevacizumab injections within 1 year in the

637ELATED MACULAR DEGENERATION

FIGURE 3. Course of distance visual acuity (VA [logMAR]), reading acuity (logRAD score), and contrast sensitivity (CS score)during monthly-treated subfoveal neovascular age-related macular degeneration for 1 year. The mean improvement for distanceVA from baseline to 1 year was 8.1 ± 12.7 letters, and reading acuity increased by 0.15 ± 0.54 logRAD score units. Maximum readingspeed increased by 3.65 ± 5.17 wpm and contrast sensitivity by 0.17 ± 0.18 CS score units. Evaluated by a random slope and randomintercept model, distance visual acuity and contrast sensitivity improvement was statistically significant. Reading acuity showed nostatistically significant 1-year function gain (see also Tables 1 and 3).

FIGURE 4. Course of number of normal function loci (NFL), mild and severe relative scotoma (MR-SC and SR-SC, respectively),and absolute scotoma (A-SC) during 1 year of monthly anti–vascular endothelial growth factor treatment in patients with subfovealneovascular age-related macular degeneration. The number of normal function loci test points changed from 15% ± 21% to 40% ±32% after 1 year of continuous treatment. Accordingly, the number of absolute scotoma decreased from 32% ± 16% to 18% ± 24%.In accordance with the lowest percentage of absolute scotoma at month 11, normal function loci were highest at month 11. Percentageof mild relative scotoma remained stable over the observational period (Table 2).

Avastin (bevacizumab) for choroidal neovascularization(ABC) trial gained 4 contrast-sensitivity letters, equivalentto a mean increase of 0.1 log contrast sensitivity.27 Thepatients who were treated monthly in the Inhibit VEGF

638 AMERICAN JOURNAL OF

in age-related choroidal neovascularisation (IVAN) studygained a mean of 2.3 contrast-sensitivity letters, whereasthose who were treated as needed (pro re nata [PRN])gained a mean of 1.74 contrast-sensitivity letters,

OCTOBER 2013OPHTHALMOLOGY

TABLE 3. Change of Visual Function Variables Over 1 Year in Patients With Neovascular Age-Related Macular Degeneration UnderMonthly Standard Anti–Vascular Endothelial Growth Factor Treatment Evaluated by a Random-Slope and Random-Intercept Model

With an Unstructured Covariance Matrix

logRAD Score Reading Speed (wpm)

Effect Estimate [95% CI] P Value Estimate [95% CI] P Value

Time �0.005 [�0.01; 0.005] .31 �0.04 [�1.2; 1.11] .94

Age �0.009 [�0.005; 0.025] .21 �0.37 [�2.0; 1.2] .65

Type II vs type I 0.2 [�0.09; 0.49] .17 �29.18 [�60.3; 1.91] .065

Contrast Sensitivity Score logMAR

Effect Estimate [95% CI] P Value Estimate [95% CI] P Value

Time 0.011 [0.007; 0.015] <.0001 �0.007 [�0.01; �0.002] .011

Age �0.008 [�0.01; �0.002] .0046 0.008 [�0.001; 0.016] .07

Type II vs type I 0.015 [�0.09; 012] .78 0.08 [�0.08; 0.25] .29

CI ¼ confidence interval; logMAR ¼ logarithm of minimal angle of resolution; wpm ¼ words per minute.

The table presents the improvements of distance visual acuity (logMAR), reading acuity (logRADscore), reading speed (wpm), and contrast

sensitivity over a monthly 1-year observational period. The estimated effect [95%CI] of each independent variable (age, lesion type, and obser-

vational period [time]) on the respective visual function variable is shown. Type I ¼ occult and minimal classic lesion type; type II¼ classic and

predominantly classic lesion type.

TABLE 4. Change of Mean Central Retinal Sensitivity Over

1 Year According to Time, Age, and Lesion Type in PatientsWith Neovascular Age-Related Macular Degeneration Under

Monthly Standard Anti–Vascular Endothelial Growth Factor

Treatment Evaluated by a Random-Slope and Random-

Intercept Model With an Unstructured Covariance Matrix

Mean CRS

Effect Estimate [CI] P Value

Time 0.21 [0.13; 0.3] <.0001

Age �0.14 [�0.26; �0.02] .02

Type II vs type Ia �0.47 [�2.77; 1.83] .68

CI ¼ confidence interval; CRS ¼ central retinal sensitivity.aType I ¼ occult and minimal classic lesion type; type II ¼

classic and predominantly classic lesion type.

TABLE 5. Change of Absolute Scotoma Size Over 1 Year

According to Time, Lesion Type, Lesion Size, and Age inPatients With Neovascular Age-Related Macular

Degeneration Under Monthly Standard Anti–Vascular

Endothelial Growth Factor Treatment Evaluated by

a Random-Slope and Random-Intercept Model With anUnstructured Covariance Matrix

Absolute Scotoma Size (mm2)

Effect Estimate [CI] P Value

Time �0.21 [�0.43; 0.003] .053

Age 0.18 [�0.04; 0.41] .11

Type II vs type I 1.3 [�2.35; 4.98] .47

Lesion size 0.5 [0.22; 0.85] .0015

Type I ¼ occult and minimal classic lesion type; type II ¼classic and predominantly classic lesion type.

suggesting that monthly treatment results in greater benefitfor patients’ contrast sensitivity. Mean retinal sensitivityalso improved in our study, supporting findings in previousstudies in neovascular AMD using a PRN treatmentregimen.6,28 One of these earlier studies found that meanretinal sensitivity increased from 4.89 dB to 9.82 dB.28

Patients in our study had higher baseline values than inthe other studies, with a mean central retinal sensitivityof 7.3 dB, and we recorded a lower mean increase of3.5 dB. Thus, the increase in retinal sensitivity could berelated to divergent baseline values and different treatmentregimens. However, in any event, irrespective of treatmentregimen, contrast sensitivity and retinal sensitivity seem toimprove under anti-VEGF treatment; thus, they seem to bereliable markers for the effect of anti-VEGF.

VOL. 156, NO. 4 FUNCTIONAL RECOVERY IN WET AGE-R

Contrary to contrast sensitivity and mean retinal sensi-tivity, reading acuity and maximum reading speed failedto improve in our study. These findings are consistentwith those of a recent study that found no improvementin the reading acuity of 137 patients with newly diagnosedneovascular AMD 12 months after initiation of anti-VEGFtreatment. These patients received PRN treatment andwere evaluated for vision-related quality of life, readingacuity, and distance VA at baseline and at 6 and 12 months(Finger R, et al. IOVS 2012;547:ARVO E-Abstract 6521).However, 1 study in patients with neovascular AMD

found a statistically significant improvement in quality oflife and near VA during 3 months of treatment and anotherin reading acuity over 3 months.1,5 Although we found

639ELATED MACULAR DEGENERATION

TABLE 6. Correlations of Visual Function Courses in Patients With Neovascular Age-Related Macular Degeneration Under MonthlyStandard Anti–Vascular Endothelial Growth Factor Treatment

logRAD Score logMAR Wpm CS Score Mean CRS (dB)

Reading acuity (logRAD score) - r ¼ 0.616

P < .0001

r ¼ �0.680

P < .0001

r ¼ �0.331

P ¼ .009

r ¼ �0.118

P ¼ .37

Distance VA (logMAR) - - r ¼ �0.241

P ¼ .062

r ¼ �0.389

P ¼ .0019

r ¼ �0.089

P ¼ .49

Maximum reading speed (wpm) - - - r ¼ 0.257

P ¼ .045

r ¼ 0.116

P ¼ .37

Contrast-sensitivity (CS score) - - - - r ¼ 0.134

P ¼ .30

CRS ¼ central retinal sensitivity; CS ¼ contrast sensitivity; logMAR ¼ logarithm of minimal angle of resolution; wpm ¼ words per minute.

The individual slopes of visual function variables were evaluated for each patient over time. The slopes of reading acuity (logRAD score),

maximum reading speed (wpm), distance visual acuity (logMAR), contrast sensitivity (CS score), and mean central retinal sensitivity measured

in dBwere correlated by Pearson correlation. Thus, the improvement courses of different variables were compared. The slope of reading acuity

correlated moderately with distance visual acuity and reading speed, respectively. The slopes of all the other variables showed only a weak or

no significant correlation.

FIGURE 5. To compare the individual courses and improvements of each visual function variable, each individual slope of the func-tional parameters was correlated to each other. Presented here, for example, is the correlation of the individual slopes of reading acuity(logRAD score) and distance visual acuity (logMAR). The correlation coefficient was 0.615, P< .0001, which revealed a moderatecorrelation. The correlations of the other visual function tests showed only a weak or no association. Individual correlations of theslopes of the functional parameters can be found in Table 6.

a short-term improvement in reading acuity at 3 months(Figure 3), this improvement was not maintained for therest of the observation period. Obviously, standard anti-VEGF increases reading acuity in the short term, butbecause of the strong intervisit variations and individualfluctuation range this improvement cannot be maintainedin the long run.5,33 Further, a central scotoma has beenshown to reduce reading speed and reading ability andscotoma size has been suggested to correlate with readingability.34–36 Thus, probably the lack of decrease in

640 AMERICAN JOURNAL OF

absolute scotoma size and the lack of improvement infixation stability in this study can explain the limits ofreading acuity improvement because fixation and absolutescotoma size are an important prerequisite for readingability and reading speed.9,34–38 This implicates onceagain that reading acuity and reading speed seem moresensitive to the status of scotoma size and fixation stabilitythan distance visual acuity. However, this also implicatesthat patients demonstrating significant improvement infixation stability and scotoma size reduction may also

OCTOBER 2013OPHTHALMOLOGY

show improvement in reading acuity. Hence, further studiesare needed to confirm or refute our findings.

In contrast to the stagnant fixation stability in our study,1 study using a PRN treatment regimen demonstrateda statistically significant increase in fixation stability within12 months,28 and another in which only 15 patients withneovascular AMD were treated with intravitreal anti-VEGF for 3 months also found such an improvement.39

These discrepancies with our study results are possiblyrelated to the lower mean retinal sensitivity at baselinein the earlier studies, to the greater retinal sensitivityincreases compared with our study, and to the stagnantscotoma area despite monthly treatment in our study.Furthermore, the aforementioned study evaluated fixationstability based on a bivariate contour ellipse area and notwith the MP-1 software. Clearly, continuous longitudinalstudies are still needed to evaluate the fixation stabilitywith regard to the location, preferred retinal locus, andthe treatment regimen.

Besides fixation stability, scotoma size also seems to be animportant factor in the ability to read.34,35 Our studypopulation lacked a significant decrease in absolutescotoma size during the observational period. This is incontrast to previous published data showing a significantabsolute scotoma decrease after 12 months.28 However,in this aforementioned study, simple Student t tests wereperformed, comparing 3-, 6-, and 12-month results withthe baseline data. By contrast, we conducted an elaboratemixed-model analysis including monthly data and addi-tional influencing variables such as lesion type, age, andlesion size. Analyzing our current data with a Studentt test, comparing baseline scotoma size with scotoma sizeafter 12 months, would also result in highly significantscotoma size decrease. Further, our calculation modelgave a P value of .053, implying that a greater sample sizemay result in significant reduction, even in a dedicatedmodel adjusting for age, lesion size, and lesion type.Thus, results should always be interpreted with respect totheir statistical evaluation, as adjusting for several factorsin multimodal models may not result in the same level ofsignificance, but may help to identify key aspects of retinalfunction and treatment response.

Lesion size was one of the above-mentioned confoundingfactors in the evaluation of absolute scotoma size, whichpredicted the decrease in scotoma area. Previous studiescould already demonstrate that larger CNV lesions andleakage size were associated with vision loss in neovascularAMD patients under anti-VEGF treatment.40,41 In theseprevious studies, CNV lesion size, but not CNV lesiontype, was predictive for the visual outcome. Concordantly,

VOL. 156, NO. 4 FUNCTIONAL RECOVERY IN WET AGE-R

the reduction in absolute scotoma area did not differ withCNV lesion type in our study. Also retinal sensitivity,distance VA, contrast sensitivity, and reading acuity didnot differ according to the CNV lesion type, which hasalso been demonstrated in previous studies.28,42,43 Toconclude, CNV type does not seem to influence visualfunction increase under anti-VEGF therapy, whereas lesionsize at baseline seems to be a factor influencing the course ofscotoma area and visual function.Last but not least, the individual slopes of each visual

function parameter were compared to evaluate whetherthe course of distance VA resembles the course of theremaining visual function parameters under anti-VEGFtreatment. Actually, the slopes of distance VA and meanretinal sensitivity did not correlate at all and the correla-tion of the slopes of contrast sensitivity and distance VAwere only weak. A previous cross-sectional study demon-strated that mean retinal sensitivity is not associated withdistance VA, and another study showed that retinal sensi-tivity improved further when distance VA had alreadystabilized.44,45 In the Radiation Therapy for Age-relatedMacular Degeneration (RAD) study, in turn, patients losta mean of 22 letters over 2 years in distance VA, butonly 9 contrast-sensitivity letters,46 implicating thatcompared with the changes in distance VA, the changesin contrast sensitivity were quite limited. Interestingly,the slopes of distance VA and reading acuity had the high-est correlation coefficient among the visual function vari-ables that we investigated, although the correlation wasonly moderate. Previous studies in neovascular AMDassumed a 0.3 log unit difference between near and distanceVA measurements, with the highest discrepancy betweendistance VA, contrast sensitivity, and reading performancein the mid range of vision decrease, suggesting that readingacuity decreases to a greater extent than distance VA andcontrast sensitivity.33 Other investigators have concludedthat reading acuity is more severely impaired and presentswith a greater range of improvement than distance VA.1,4

To subsume, all these findings implicate that each visualfunction parameter seems to have its individual recoverypattern under anti-VEGF treatment; thus the course ofdistance VA is only partially representative.To summarize, although distanceVA is a well-established

and important tool to assess the efficacy of new therapies inclinical trials, it does not seem to comprehensively reflectoverall visual function gain or improvement in the differentvision-related tasks of everyday life. Additional visual func-tion variables should be examined in routine clinical prac-tice as well as in multicenter trials to evaluate patients’subjective vision impairment or improvement.

ALL AUTHORSHAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSUREOF POTENTIAL CONFLICTS OF INTEREST.Ursula Schmidt-Erfurth receives consultancy and lecture fees and travel support fromAlcon Laboratory Inc, Bayer Health Care, and Novartis. The authorsindicate no funding or financial support. Contributions of authors: design and conduct of the study (M.M., C.K., W.H., C.S., U.S.E.), collection andmanagement of the data (M.M., F.S., P.R., T.M., A.M., C.S., S.S.), analysis and interpretation of the data (M.M., W.H., T.M., S.S., C.S.), preparationof the manuscript (M.M.), and review of the manuscript (M.M, C.K, C.S, W.H., F.S., U.S.E).

641ELATED MACULAR DEGENERATION

The authors thank Professor Lee Jampol (Northwestern University, Chicago, Illinois) and Dr Rene Ruckert (Scicograde, Basel, Switzerland) for theircareful review of the manuscript. They also thank Drs Ramzi Sayegh and Marion Funk (Department of Ophthalmology, Medical University Vienna,Austria) for performing the intravitreal injections. The authors have obtained written permission from all persons named in the acknowledgment.

REFERENCES

1. Frennesson C, Nilsson UL, Peebo BB, Nilsson SE. Significantimprovements in near vision, reading speed, central visualfield and related quality of life after ranibizumab treatmentof wet age-related macular degeneration. Acta Ophthalmol2010;88(4):420–425.

2. Bressler SB. Health maintenance issues of the elderly. Vision:age-related macular degeneration. Md Med J 1989;38(2):135–137.

3. Hazel CA, Petre KL, Armstrong RA, Benson MT, Frost NA.Visual function and subjective quality of life compared insubjects with acquired macular disease. Invest Ophthalmol

Vis Sci 2000;41(6):1309–1315.4. Kiss CG, Barisani-Asenbauer T, Maca S, Richter-Mueksch S,

Radner W. Reading performance of patients with uveitis-associated cystoid macular edema. Am J Ophthalmol 2006;142(4):620–624.

5. Munk M, Kiss C, Huf W, et al. Therapeutic interventions formacular diseases show characteristic effects on near anddistance visual function [published online ahead of print].Retina. doi: 10.1097/IAE.0b013e318285cc0c.

6. Parravano M, Oddone F, Tedeschi M, et al. Retinal func-tional changes measured by microperimetry in neovascularage-related macular degeneration patients treated with rani-bizumab. Retina 2009;29(3):329–334.

7. Mones J, Rubin GS. Contrast sensitivity as an outcomemeasure in patients with subfoveal choroidal neovascularisa-tion due to age-related macular degeneration. Eye 2005;19(11):1142–1150.

8. Stifter E, Sacu S, Benesch T, Weghaupt H. Impairment ofvisual acuity and reading performance and the relationshipwith cataract type and density. Invest Ophthalmol Vis Sci

2005;46(6):2071–2075.9. Legge GE, Ross JA, Isenberg LM, LaMay JM. Psychophysics

of reading. Clinical predictors of low-vision reading speed.Invest Ophthalmol Vis Sci 1992;33(3):677–687.

10. Rubin GS, West SK, Munoz B, et al. A comprehensiveassessment of visual impairment in a population of olderAmericans. The SEE Study. Salisbury Eye Evaluation Project.Invest Ophthalmol Vis Sci 1997;38(3):557–568.

11. Richter-Mueksch S, Stur M, Stifter E, Radner W. Differencesin reading performance of patients with Drusen maculopathyand subretinal fibrosis after CNV. Graefes Arch Clin ExpOphthalmol 2006;244(2):154–162.

12. Cacho I, Dickinson CM, Smith HJ, Harper RA. Clinicalimpairment measures and reading performance in a largeage-related macular degeneration group. Optom Vis Sci2010;87(5):344–349.

13. Roesel M, Heimes B, Heinz C, Henschel A, Spital G,Heiligenhaus A. Comparison of retinal thickness andfundus-related microperimetry with visual acuity in uveiticmacular oedema. Acta Ophthalmol 2011;89(6):533–537.

14. Keane PA, Patel PJ, Ouyang Y, et al. Effects of retinalmorphology on contrast sensitivity and reading ability in

642 AMERICAN JOURNAL OF

neovascular age-related macular degeneration. InvestOphthalmol Vis Sci 2010;51(11):5431–5437.

15. Hatef E, Colantuoni E, Wang J, et al. The relationshipbetween macular sensitivity and retinal thickness in eyeswith diabetic macular edema. Am J Ophthalmol 2011;152(3):400–405.e2.

16. Ferris FL 3rd, Kassoff A, Bresnick GH, Bailey I. New visualacuity charts for clinical research. Am J Ophthalmol 1982;94(1):91–96.

17. Radner W, Obermayer W, Richter-Mueksch S, Willinger U,Velikay-Parel M, Eisenwort B. The validity and reliability ofshort German sentences for measuring reading speed.Graefes

Arch Clin Exp Ophthalmol 2002;240(6):461–467.18. Burggraaff MC, van Nispen RM, Hoek S, Knol DL, van

Rens GH. Feasibility of the Radner Reading Charts in low-vision patients. Graefes Arch Clin Exp Ophthalmol 2010;248(11):1631–1637.

19. Radner W,Willinger U, Obermayer W, Mudrich C, Velikay-Parel M, Eisenwort B. [A new reading chart for simultaneousdetermination of reading vision and reading speed]. KlinMonbl Augenheilkd 1998;213(3):174–181.

20. Powers MK. Paper tools for assessing visual function. Optom

Vis Sci 2009;86(6):613–618.21. Prager F, Michels S, Simader C, Geitzenauer W, Schmidt-

Erfurth U. Changes in retinal sensitivity in patients withneovascular age-related macular degeneration after systemicbevacizumab (avastin) therapy. Retina 2008;28(5):682–688.

22. Kriechbaum K, Prager F, Geitzenauer W, et al. Association ofretinal sensitivity and morphology during antiangiogenictreatment of retinal vein occlusion over one year. Ophthal-mology 2009;116(12):2415–2421.

23. Fujii GY, De Juan E Jr, HumayunMS, Sunness JS, Chang TS,Rossi JV. Characteristics of visual loss by scanning laserophthalmoscope microperimetry in eyes with subfovealchoroidal neovascularization secondary to age-relatedmacular degeneration. Am J Ophthalmol 2003;136(6):1067–1078.

24. Fujii GY, de Juan E Jr, Sunness J, HumayunMS, Pieramici DJ,Chang TS. Patient selection for macular translocationsurgery using the scanning laser ophthalmoscope. Ophthal-mology 2002;109(9):1737–1744.

25. McCulloch CE, Neuhaus JM. Prediction of random effects inlinear and generalized linear models under model misspecifi-cation. Biometrics 2011;67(1):270–279.

26. Bansback N, Czoski-Murray C, Carlton J, et al. Determinantsof health related quality of life and health state utility inpatients with age related macular degeneration: the associa-tion of contrast sensitivity and visual acuity. Qual Life Res

2007;16(3):533–543.27. Patel PJ, Chen FK, Da Cruz L, Rubin GS, Tufail A. Contrast

sensitivity outcomes in the ABC Trial: a randomized trial ofbevacizumab for neovascular age-related macular degenera-tion. Invest Ophthalmol Vis Sci 2011;52(6):3089–3093.

28. Cho HJ, Kim CG, Yoo SJ, et al. Retinal functional changesmeasured by microperimetry in neovascular age-related

OCTOBER 2013OPHTHALMOLOGY

macular degeneration treated with ranibizumab. Am JOphthalmol 2013;155(1):118–126.e111.

29. Chakravarthy U, Harding SP, Rogers CA, et al. Ranibizumabversus bevacizumab to treat neovascular age-related maculardegeneration: one-year findings from the IVAN randomizedtrial. Ophthalmology 2012;119(7):1399–1411.

30. Martin DF, Maguire MG, Ying GS, Grunwald JE, Fine SL,Jaffe GJ. Ranibizumab and bevacizumab for neovascularage-related macular degeneration. N Engl J Med 2011;364(20):1897–1908.

31. Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab forneovascular age-related macular degeneration. New Engl JMed 2006;355(14):1419–1431.

32. Brown DM, Kaiser PK, Michels M, et al. Ranibizumab versusverteporfin for neovascular age-related macular degeneration.New Engl J Med 2006;355(14):1432–1444.

33. Hogg RE, Chakravarthy U. Visual function and dysfunctionin early and late age-related maculopathy. Prog Retin EyeRes 2006;25(3):249–276.

34. Ergun E,MaarN, RadnerW, Barbazetto I, Schmidt-ErfurthU,SturM. Scotoma size and reading speed in patients with subfo-veal occult choroidal neovascularization in age-relatedmacular degeneration. Ophthalmology 2003;110(1):65–69.

35. Rayner K. Eye guidance in reading: fixation locations withinwords. Perception 1979;8(1):21–30.

36. McMahon TT, Hansen M, Viana M. Fixation characteristicsin macular disease. Relationship between saccadic frequency,sequencing, and reading rate. Invest Ophthalmol Vis Sci 1991;32(3):567–574.

37. Whittaker SG, Lovie-Kitchin J. Visual requirements forreading. Optom Vis Sci 1993;70(1):54–65.

38. Crossland MD, Culham LE, Rubin GS. Predicting readingfluency in patients with macular disease. Optom Vis Sci

2005;82(1):11–17.

VOL. 156, NO. 4 FUNCTIONAL RECOVERY IN WET AGE-R

39. Gonzalez EG, Tarita-Nistor L, Mandelcorn ED,Mandelcorn M, Steinbach MJ. Fixation control before andafter treatment for neovascular age-related macular degener-ation. Invest Ophthalmol Vis Sci 2011;52(7):4208–4213.

40. Rosenfeld PJ, Shapiro H, Tuomi L, Webster M, Elledge J,Blodi B. Characteristics of patients losing vision after 2 yearsof monthly dosing in the phase III ranibizumab clinical trials.Ophthalmology 2011;118(3):523–530.

41. Bloch SB, la Cour M, Sander B, et al. Predictors of 1-yearvisual outcome in neovascular age-related macular degenera-tion following intravitreal ranibizumab treatment. ActaOphthalmol 2013;91(1):42–47.

42. Boyer DS, Antoszyk AN, Awh CC, Bhisitkul RB, Shapiro H,Acharya NR. Subgroup analysis of the MARINA study ofranibizumab in neovascular age-related macular degenera-tion. Ophthalmology 2007;114(2):246–252.

43. Munk M, Kiss C, Sulzbacher F, et al. Short-term progressionof wet AMD and correlation with 1-year treatment results.Acta Ophthalmol 2012;90(6):e420–e427.

44. Kiss CG, Geitzenauer W, Simader C, Gregori G, Schmidt-Erfurth U. Evaluation of ranibizumab-induced changes inhigh-resolution optical coherence tomographic retinalmorphology and their impact on visual function. Invest

Ophthalmol Vis Sci 2009;50(5):2376–2383.45. Parravano M, Oddone F, Tedeschi M, et al. Retinal func-

tional changes measured by microperimetry in neovascularage-related macular degeneration treated with ranibizumab:24-month results. Retina 2010;30(7):1017–1024.

46. Bellmann C, Unnebrink K, Rubin GS, Miller D, Holz FG.Visual acuity and contrast sensitivity in patients with neovas-cular age-related macular degeneration. Results from theRadiation Therapy for Age-Related Macular Degeneration(RAD-) Study. Graefes Arch Clin Exp Ophthalmol 2003;241(12):968–974.

643ELATED MACULAR DEGENERATION

Biosketch

Marion Munk, MD, PhD is a resident at the Medical University Vienna. She recently finished her PhD emphasizing on

visual function and retinal morphology. She is an ophthalmologist with a focus on clinical research in retina, age-

related macular degeneration and cystoid macular edema applying state of the art imaging as optical coherence

tomography and functional magnetic resonance imaging. She is an investigator in many global clinical trials, and avid

teacher and lecturer. Dr Munk is currently pursuing a temporary research fellowship at the Northwestern University,

Chicago, Illinois.

643.e1 OCTOBER 2013AMERICAN JOURNAL OF OPHTHALMOLOGY

SUPPLEMENTAL FIGURE. Course of absolute scotoma area in patients with neovascular age-related macular degeneration treatedwith monthly anti–vascular endothelial growth factor (VEGF) with respect to choroidal neovascularization lesion type (type I vs typeII). Microperimetry test points with 0 dB were considered as absolute scotoma and absolute scotoma area was then evaluated throughinterpolation by dedicated software. Therefore, each stimulation locus presenting with 0 dBwas multiplied with its associated area forthe evaluation of the size of severe relative scotoma. The change in absolute scotoma area did not differ between type I and type IIlesions (P [ .47), which is also apparent by the overlapping confidence intervals (CI). Type I [ occult and minimal classic lesiontype; Type II [ classic and predominantly classic lesion type.

VOL. 156, NO. 4 643.e2FUNCTIONAL RECOVERY IN WET AGE-RELATED MACULAR DEGENERATION