Evaluation of Macular Function Using Focal Macular...

Evaluation of Macular Function Using Focal MacularElectroretinography in Eyes with Macular EdemaAssociated with Branch Retinal Vein Occlusion

Ken Ogino, Akitaka Tsujikawa, Tomoaki Murakami, Yuki Muraoka,Yumiko Akagi-Kurashige, Kenji Ishihara, Kazuaki Miyamoto, Hajime Nakamura,and Nagahisa Yoshimura

PURPOSE. To study the usefulness of focal macular electroreti-nography (fmERG) for evaluation of macular function in eyeswith macular edema (ME) associated with branch retinal veinocclusion (BRVO).

METHODS. The authors prospectively performed fmERG on 34patients with untreated unilateral BRVO at the initial visit and12 months later. Amplitudes and latencies of the a-wave, b-wave, and photopic negative response (PhNR) were comparedwith visual acuity, with retinal sensitivity measured by micro-perimetry, and with measurements obtained by optical coher-ence tomography.

RESULTS. In eyes with ME from BRVO, amplitudes of the a-wave,b-wave, and PhNR were reduced significantly and latencieswere prolonged significantly compared with those of healthyfellow eyes (P � 0.05). Relative amplitudes and latencies werenot correlated with visual acuity but, rather, tended to becorrelated with retinal sensitivity within the macular area.Among all parameters studied by fmERG, relative amplitude ofthe PhNR was most strongly correlated with central fovealthickness (r � �0.465; P � 0.007). In addition, the height ofthe serous retinal detachment showed a correlation with therelative amplitude of the PhNR (r � �0.376; P � 0.034). At 12months, the amplitude of the b-wave and the PhNR improvedsignificantly, in parallel with resolution of the ME (P � 0.015;P � 0.033).

CONCLUSIONS. In eyes with ME from BRVO, amplitudes andlatencies seen by fmERG were correlated with other biologicalparameters. Based on findings of the present study, fmERGappears to be useful as a functional examination within themacular area affected by BRVO. (Invest Ophthalmol Vis Sci.2011;52:8047–8055) DOI:10.1167/iovs.11-8143

Macular edema (ME) is one of the most vision-threateningcomplications associated with branch retinal vein occlu-

sion (BRVO). Since the Branch Vein Occlusion Study Groupreported the efficacy of grid laser photocoagulation, it has beenrecognized as the standard treatment for ME from BRVO,1 but,

recently, an increasing number of reports have shown theefficacy of newer treatments on ME from BRVO, such as intra-vitreal injection of triamcinolone acetonide, bevacizumab, or,ranibizumab.2–4 Generally, both the severity of the ME and theeffect of the treatment have been evaluated by quantitativemeasurement of foveal thickness using optical coherence to-mography (OCT).5,6 As a functional parameter, we typicallyuse visual acuity (VA) measurement. Although ME associatedwith BRVO usually involves the larger macular area, VA mea-surement reflects only foveal function. In addition, physicianssometimes note a discrepancy between the morphologic se-verity of ME and VA. To evaluate the severity of ME and itsresponse to treatment, it is essential to establish another func-tional examination that reflects not only the fovea but, also, thelarger macular area.7,8

Full-field electroretinography (ffERG), which allows exami-nation of the entire retinal function, is used widely as a fun-ctional examination in various diseases.9 Using ffERG, Chenet al.10 reported an abnormal photopic negative response(PhNR), which is thought to reflect the function of the innerretina in eyes with BRVO. In contrast to ffERG, the focalmacular ERG (fmERG) allows examination of retinal functionwithin only the macular area.11,12 Using fmERG, Terasaki etal.13,14 demonstrated a correlation between increased fovealthickness from diabetic ME and decreased macular function.

Thus far, as a functional measurement of a focal area withinthe macula, multifocal ERG has been used in various retinaldiseases,15–17 especially retinal dystrophies.18–20 Comparedwith multifocal ERG, fmERG allows more reliable recording ofthe whole macular area, even from patients with poor fixationor poor VA, by monitoring the fundus through an infraredcamera and manually adjusting the stimulus at the fo-vea.13,14,21–25 We hypothesize that fmERG might enable us toevaluate more effectively the impaired macular function due toME secondary to BRVO. Little information, however, is avail-able about macular function examined by fmERG in eyes withBRVO. In the study described herein, fmERG in eyes with MEassociated with BRVO was performed during the acute phaseand at 12 months, at which time ME was resolved in most eyes,and we studied the correlations of fmERG with other biologicalparameters. Based on our findings, we evaluated the usefulnessof fmERG to better define macular function of these eyes.

PATIENTS AND METHODS

This prospective study consisted of 34 patients with ME secondary tountreated unilateral BRVO who made their initial visit to Kyoto Uni-versity Hospital between May 2009 and March 2010. This study wasapproved by the Institutional Review Board at Kyoto University Grad-uate School of Medicine and adhered to the tenets of the Declaration

From the Department of Ophthalmology and Visual Sciences,Kyoto University Graduate School of Medicine, Kyoto, Japan.

Submitted for publication June 29, 2011; revised August 24, 2011;accepted September 4, 2011.

Disclosure: K. Ogino, None; A. Tsujikawa, None; T. Mu-rakami, None; Y. Muraoka, None; Y. Akagi-Kurashige, None; K.Ishihara, None; K. Miyamoto, None; H. Nakamura, None; N.Yoshimura, None

Corresponding author: Akitaka Tsujikawa, Department of Oph-thalmology, Kyoto University Graduate School of Medicine, Sakyo-ku,Kyoto 606-8507, Japan; [email protected].

Retina

Investigative Ophthalmology & Visual Science, October 2011, Vol. 52, No. 11Copyright 2011 The Association for Research in Vision and Ophthalmology, Inc. 8047

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933458/ on 05/15/2018

of Helsinki. Written informed consent was obtained from each patient.Patients with coexisting ocular disease (epiretinal membrane, glau-coma, diabetic retinopathy, or senile cataract that resulted in poor-quality OCT images and fmERG) in either eye were excluded from thepresent study. Eyes with hemi-central retinal vein occlusion (hemi-CRVO) were also excluded from the present study. The diagnosis ofBRVO was based on fundus examination and fluorescein angiographyperformed by two retina specialists (AT, TM). At the initial visit, eachpatient underwent complete ophthalmic examination, including best-corrected VA measurement, slit-lamp biomicroscopy, indirect fundusophthalmoscopy, OCT measurement, fluorescein angiography, micro-perimetry, and fmERG. At the scheduled visit at 12 months, eachpatient underwent another complete examination, including best-cor-rected VA measurement, slit-lamp biomicroscopy, indirect fundus oph-thalmoscopy, OCT measurement, and fmERG. During the study period,persistent ME was treated with grid photocoagulation in 11 (32%) eyesand with pars plana vitrectomy in 4 (12%) eyes.

Best-corrected VA was measured with a Landolt chart and wasconverted to a logarithm of the minimum angle of resolution (logMAR).

Fluorescein angiography was performed using a confocal laser scan-ning system (HRA-2; Heidelberg Engineering, Heidelberg, Germany).At each scheduled visit, the entire macular area was examined withmultimodality imaging (Spectralis�OCT; Heidelberg Engineering). Us-ing a vertical sectional image centered on the fovea obtained at theinitial visit and at 12 months, we performed three measurements in thefovea; these consisted of center point thickness, height of the serousretinal detachment, and sensory retinal thickness. Center point thick-ness was defined as the distance between the internal limiting mem-brane and the retinal pigment epithelium (RPE) at the center of thefovea; height of serous retinal detachment was defined as the distancebetween the RPE and the bottom of the detached neurosensory retina,just beneath the fovea; sensory retinal thickness was calculated bysubtracting the height of the serous retinal detachment from the centerpoint thickness.

In 31 eyes with BRVO, retinal sensitivity within the macular areawas examined with a fundus-monitored microperimeter (Micro Perim-eter 1 [MP1]; Nidek, Gamagori, Japan). A 4–2-staircase strategy withGoldmann III size stimuli was used, and 57 stimulus locations withinthe central 10° were examined by microperimetry. Each stimulus waslocated according to the measurement points in Humphrey 10–2, withsome additional points (Fig. 1A). The white background illuminationwas set at 1.27 cd/m2. The differential luminance, defined as thedifference between stimulus luminance and background luminance,was 1.27 cd/m2 at 0 dB stimulation, and the maximum stimulus atten-uation was 20 dB. The duration of the stimulus was 200 ms, and thefixation target varied in size (2° cross for central fixation, 4° or 6° crossfor paracentral fixation) according to the VA of the individual patient.There were 17 and 37 measurement points within the central 4° and 8°areas.

Recording of fmERG has been reported in detail previously.26

Briefly, after the pupils of both eyes were maximally dilated, a Burian-Allen bipolar contact lens electrode (Hansen Ophthalmic Laboratories,Iowa city, IA) was placed in the conjunctival sac of each eye undertopical anesthesia. A chloride silver electrode was attached to the leftearlobe as a ground electrode. fmERG was elicited by 15° circularstimuli positioned on the macular area, using a prototype of the ER-80(Kowa, Tokyo, Japan), which was composed of an infrared camera(Kowa) and a stimulation system (Mayo Co., Nagoya, Japan). Theluminances of white stimulus light and background illumination were181.5 and 6.9 cd/m2, respectively. The 15° circular stimulus wascarefully and constantly centered on the fovea, as observed throughthe infrared camera (Fig. 1B). The fmERG was recorded with 2-Hzrectangular stimuli (150 ms with the light on and 350 ms with the lightoff). Affected eyes were examined before fellow eyes. The recording(100–150 responses) was made twice to confirm reproducibility, anda total of 200 to 300 responses were averaged by the signal processor(Neuropack MEB-2204; Nihon Kohden, Tokyo, Japan). The fmERGresponse was digitized at 10 kHz with a band-pass filter of 5 to 500 Hz

FIGURE 1. Measurement area with MP1 and fmERG. (A) Fifty-sevenlocations covering the central 10° were examined with the MP1 tomeasure the retinal sensitivity of the macular area. Seventeen measure-ment points were located within the central 4° (yellow circle), and 37points were located within the central 8° (red circle). (B) fmERG waselicited by 15° circular stimuli positioned on the fovea. (C) fmERGobtained from a normal fellow eye. A total of 200 to 300 responseswere averaged by a signal processor. Black arrowhead: beginning ofstimulation; red arrows: amplitudes of the waves.

FIGURE 2. Amplitudes (A) and laten-cies (B) of 15° circular fmERG ob-tained from eyes with ME associatedwith BRVO and from fellow eyes atthe initial visit. Error bar, SD. *P �0.05, †P � 0.01 in paired t-test.

8048 Ogino et al. IOVS, October 2011, Vol. 52, No. 11


for the a-wave, b-wave, and PhNR and with a band-pass filter of 50 to500 Hz for oscillatory potentials. The amplitudes of the a-wave andb-wave were measured, respectively, from the baseline to the peak ofthe a-wave and from the trough of the a-wave to the peak of the b-wave(Fig. 1C). The amplitude of PhNR was measured from the peak of theb-wave to the trough of PhNR, based on our previous reports (Fig.1C).26–28 Latency was defined as the time from the beginning ofstimulation to the peak of each component.

Statistical analysis was performed (PASW Statistics, version 17.0;SPSS, Chicago, IL). All values are presented as mean � SD. Paired t-testwith Bonferroni correction was used to compare amplitudes and la-tencies of the a-wave, b-wave, and PhNR between affected eyes andunaffected fellow eyes. The initial fmERG parameters, VA, and OCTparameters with values obtained at 12 months were also comparedwith a paired t-test with Bonferroni correction. To compare the fmERGparameters with other measurement values (VA, retinal sensitivity,OCT parameters), we used the relative amplitude and latency (affectedeye/fellow eye) and calculated the Pearson correlation coefficient.Because PhNR is a slow wave and does not have a sharp peak in eyeswith decreased function because of BRVO, the relative latency of PhNRwas not compared with the other values. P � 0.05 was consideredstatistically significant.

RESULTS

In the present study, we examined 34 eyes with BRVO and 34healthy fellow eyes of 34 patients (18 men, 16 women), whoranged in age from 45 to 78 years (66.8 � 9.0 years). The meanduration of symptoms was 3.8 � 7.0 months. At the initial visit,VA in logMAR fashion was 0.49 � 0.38 in affected eyes and

�0.07 � 0.13 in healthy fellow eyes. All affected eyes had MEwith cystoid spaces at the fovea, in which mean center pointthickness was 533 � 192 �m. Twenty-five (74%) of the 34affected eyes had serous retinal detachments beneath the fo-vea; the mean height of these detachments was 160 � 110 �m.Mean sensory retinal thickness was 416 � 140 �m. In thefellow eyes, OCT showed a physiologic shape of the fovea, andthe mean center point thickness in these healthy eyes was233 � 35 �m.

To evaluate the reproducibility of fmERG, the intraclasscorrelation coefficient (ICC) was calculated from the initialfmERG and from recordings obtained at 12 months from 24unaffected fellow eyes. In these unaffected fellow eyes, ampli-tudes of the a-wave, b-wave, and PhNR were 1.29 � 0.43,3.05 � 0.92, and 3.78 � 0.22 �V, respectively, at the initialvisit and 1.48 � 0.58, 3.46 � 0.87, and 3.99 � 1.04 �V, re-spectively, at 12 months. Latencies of the a-wave, b-wave, andPhNR were 21.2 � 1.3, 40.3 � 2.1, and 75.9 � 8.2 ms, respec-tively, at the initial visit and 21.1 � 1.4, 40.6 � 2.5, and75.7 � 9.1 ms, respectively, at 12 months. ICCs were 0.219,0.490, and 0.635, respectively, in amplitude and 0.445, 0.422,and 0.533, respectively, in latency of the a-wave, b-wave, andPhNR.

Of the 34 affected eyes, reliable fmERG recordings could beobtained from 32 (94%) of the eyes at the initial visit and from24 (100%) of 24 affected eyes at 12 months. In two eyes,reliable fmERG could not be obtained because of low repro-ducibility or a slanted baseline. In eyes with BRVO, a flat ERGwas not seen, but oscillatory potentials were diminished in 12eyes at the initial visit and in 11 eyes at 12 months. None of our

TABLE 1. Correlation of Parameters in Focal Macular Electroretinogram with Visual Acuity or RetinalSensitivity Examined with MP1 at the Initial Visit

Mean Retinal Sensitivity Examined with MP1 (dB)

Visual Acuity(logMAR) Center Point Within 4° Within 8°

r P r P r P r P

Relative amplitudea-Wave 0.105 0.568 0.092 0.635 0.188 0.330 0.179 0.354b-Wave �0.163 0.373 0.306 0.106 0.519 0.004 0.546 0.002PhNR �0.221 0.225 0.385 0.039 0.510 0.005 0.525 0.003

Relative latencya-Wave 0.014 0.938 �0.155 0.422 �0.351 0.062 �0.397 0.033b-Wave 0.066 0.720 �0.127 0.512 �0.279 0.142 �0.257 0.178

Pearson correlation coefficient. logMAR, logarithm of the minimum angle of resolution.

FIGURE 3. Relationship between rel-ative amplitude and retinal sensitivityin the macular area (A) or mean ret-inal thickness (B) in eyes with MEassociated with BRVO. Relative am-plitude is calculated as amplitude inthe affected eye divided by that inthe fellow eye. Mean sensitivitywithin 8° was obtained with theMP1. Center point thickness was ob-tained with a vertical section of OCT.

IOVS, October 2011, Vol. 52, No. 11 Focal Macular Electroretinography in BRVO 8049


patients reported eye pain or loss of vision after the fmERGrecording.

At the initial visit, amplitudes of the a-wave, b-wave, andPhNR in affected eyes were 0.95 � 0.39, 1.66 � 0.70, and2.22 � 1.06 �V, respectively (Fig. 2A), and were reduced sig-

nificantly compared with those of healthy fellow eyes (P �0.001). Latencies of the a-wave, b-wave, and PhNR in affectedeyes were 22.8 � 2.5, 43.8 � 3.6, and 84.0 � 11.7 ms, respec-tively (Fig. 2B), and were significantly prolonged comparedwith those in the fellow eyes (P � 0.009; P � 0.001; P �0.012).

To compare the initial fmERG parameters with other mea-surements (VA, retinal sensitivity, OCT parameters), relativeamplitudes and latencies (affected eye/fellow eye) were calcu-lated. At the initial visit, relative amplitudes and latencies werenot correlated with VA, although both relative amplitude andlatency of each wave at the initial visit showed some correla-tion with retinal sensitivity within the macular area (Table 1).Relative amplitudes of the b-wave and of PhNR had the bestcorrelation with the mean sensitivity within the central 8° area(r � 0.546, P � 0.002; r � 0.525, P � 0.003), which iscompatible with the area of fmERG stimulation (Fig. 3A).

Furthermore, we compared initial fmERG parameters withOCT measurements, also obtained at the initial visit (Table 2).Center point thickness showed a correlation with relative am-plitudes of the b-wave and PhNR. In fact among all parametersmeasured, relative amplitude of the PhNR was most stronglycorrelated with center point thickness (r � �0.465; P �0.007) (Fig. 3B). In addition, the height of the serous retinaldetachment showed a correlation with amplitudes of the b-

TABLE 2. Correlation of Parameters in fmERG with Measurements ofOCT at the Initial Visit

Center PointThickness

(�m)

SensoryRetinal

Thickness(�m)

Height ofSerousRetinal

Detachment(�m)

r P r P r P

Relative amplitudea-Wave �0.026 0.886 0.156 0.393 �0.283 0.117b-Wave �0.380 0.032 �0.227 0.211 �0.388 0.028PhNR �0.465 0.007 �0.343 0.054 �0.376 0.034

Relative latencya-Wave 0.014 0.938 0.087 0.637 0.255 0.159b-Wave 0.066 0.720 �0.059 0.750 0.291 0.106

Pearson correlation coefficient.

FIGURE 4. BRVO with extensive se-rous retinal detachment. (A) A 59-year-old man had a 1-month historyof blurred vision due to BRVO in theleft eye (0.7 on a Landolt chart,equivalent to 20/29 on a Snellenchart). (B) Retinal thickness mapshows severe ME. (C) Microperim-etry map shows decreased retinalsensitivity in the larger macula areabeyond the area affected with BRVO.(D) Initial OCT shows ME with ex-tensive serous retinal detachment.(E) ME and serous retinal detach-ment were resolved at 12 months.(F) fmERG obtained from the eye af-fected with BRVO (red, at initial visit;purple, at 12 months) and from thehealthy fellow eye (gray, at initialvisit). Black line: beginning of stim-ulation. The fmERG was severely af-fected at the initial visit and was onlypartially improved at 12 months. Rel-ative amplitude of PhNR was 0.45 atinitial visit and 0.71 at 12 months.



wave (r � �0.388; P � 0.028) and of PhNR (r � �0.376; P �0.0034). However, neither relative amplitudes nor relative la-tencies had a correlation with sensory retinal thickness. Inaffected eyes, swelling of the sensory retina did not result in areduction of amplitude or a prolongation of latency, but am-plitudes did become reduced in parallel with the height of thefoveal serous detachment. In addition, some eyes with BRVOshowed an extensive serous retinal detachment that, in somecases, extended to the unaffected side of the retina. Retinalsensitivity was substantially reduced in the area involved by theserous retinal detachment, and there was a substantial decreasein fmERG amplitude (Figs. 4, 5).

At 12 months after the initial examination, all affected eyesshowed substantial reduction in ME, and center point thick-ness was reduced to 313 � 151 �m (P � 0.001; Fig. 6). No eyehad a serous retinal detachment, but nine eyes showed residualcystoid spaces. At 12 months, VA was significantly improved to0.27 � 0.41 (P � 0.005). Amplitudes of the a-wave, b-wave,and PhNR were 0.99 � 0.55, 2.13 � 0.90, and 2.66 � 1.21 �Vin affected eyes, respectively. In affected eyes, amplitudes ofthe b-wave and PhNR were significantly improved comparedwith values at the initial visit (P � 0.015; P � 0.033) (Fig. 6).Latencies of the a-wave, b-wave, and PhNR were 22.6 � 2.8,43.6 � 3.0, and 81.5 � 11.1 ms in affected eyes, respectively;

these latencies were not significantly improved. Table 3 showsthe VA, center point thickness, and focal macular electroreti-nogram at the initial visit and at 12 months in each groupstratified by the treatment modality. Although the change offmERG parameters showed a similar tendency in each group,most were not statistically significant, perhaps because of thesmall number of eyes.

In this study, 15 affected eyes of 34 patients with BRVOshowed an area of nonperfusion that measured �5 disc diam-eters on fluorescein angiography. Table 4 shows each param-eter of the fmERG obtained at the initial visit and 12 monthslater in ischemic and nonischemic BRVO. Between ischemicBRVO and nonischemic BRVO, there were no differences inparameters of the fmERG taken at the initial visit and at 12months.

DISCUSSION

Visual prognosis in eyes with BRVO is thought to be betterthan that of eyes with CRVO.29,30 Nevertheless, some patientswith BRVO have severely impaired visual function because ofME. VA is the most common functional examination but re-flects only foveal function, and the ME caused by BRVO usually

FIGURE 5. BRVO with localized se-rous retinal detachment. (A) A 45-year-old man had a 2-month historyof blurred vision due to BRVO in theleft eye (0.9 on a Landolt chart,equivalent to 20/22 on a Snellenchart). (B) Retinal thickness mapshows severe ME. (C) Microperim-etry map shows decreased retinalsensitivity in the fovea. (D) InitialOCT shows ME with localized serousretinal detachment. (E) ME and se-rous retinal detachment are resolvedat 12 months. (F) fmERG obtainedfrom the eye affected with BRVO(red, at initial visit; purple, at 12months) and from the healthy felloweye (gray, at initial visit). Black line:beginning of stimulation. The fmERGin the affected eye was relatively pre-served both at the initial visit and at12 months. The relative amplitude ofPhNR was 0.88 at the initial visit and1.16 at 12 months.



involves the larger macular area, leading to functional impair-ment of the entire macular area. With the use of microperim-etry, Yamaike et al.8 reported a close correlation betweenretinal sensitivity in the macular area and retinal thickness ineyes with ME associated with BRVO. In the present study, weperformed fmERG and studied the correlations betweenfmERG and other biological parameters and evaluated the use-fulness of fmERG as a possible examination to better define themacular function of these eyes. fmERG allows us to performaccurate macular stimulation with monitoring of the macula

through an infrared fundus camera.12 In addition, an fmERGsystem has been commercially available since 2008, and itsclinical usefulness for several retinal diseases has been re-ported.13,14,21–25

However, it is generally recognized that amplitude of theERG does not have high reproducibility, even in subjects withhealthy eyes.31 In the present study, we used the relativeamplitudes (affected eyes/fellow eyes) to minimize this vari-ability of measurements. In addition, to confirm the reproduc-ibility of fmERG, we assessed the ICC in parameters of fmERG

FIGURE 6. Comparison of VA, retinal thickness of the fovea, and fmERG at the initial visit and at 12 months. VA was converted to logarithm ofthe minimum angle of resolution (logMAR). Center point thickness, height of the serous retinal detachment, and sensory retinal thickness wereobtained with a vertical section of OCT. *P � 0.05, †P � 0.01 in paired t-test.



with the use of fellow eyes, which were examined at both theinitial visit and at 12 months. The ICCs in amplitude andlatency of the a-wave, b-wave, and PhNR were 0.219 and 0.445,0.490 and 0.422, and 0.635 and 0.533, respectively. Based onthis assessment, we confirmed that PhNR was the most stableparameter in fmERG

In the acute phase of ME associated with BRVO, eachparameter of the fmERG showed a substantial decrease inmacular function. Although the relative amplitude and latencyof each wave were not correlated with VA, the relative ampli-tude of PhNR did show a correlation with retinal sensitivitywithin the macular area. PhNR is a negative, large, slow wavethat follows the b-wave and that typically has a larger signal/noise ratio than do other waves. Moreover, PhNR is reported toreflect inner retinal function.27,32,33 Machida et al.22,23 re-ported the usefulness of PhNR in fmERG for evaluation of theseverity of glaucoma in which ganglion cells are primarilydamaged. Because BRVO primarily causes damage in the innerretina, we hypothesized that PhNR in fmERG may be a usefulparameter with which to evaluate objectively the macularfunction in eyes with BRVO. However, there is no consensus

about how to measure the amplitude of PhNR. In previousexperiments using monkeys treated with tetrodotoxin, thePhNR trough did not reach baseline.27,28 Severe inner retinaldamage possibly resulted in PhNR trough levels above baseline.In preliminary experiments, some ischemic eyes showed sucha pattern of PhNR. In the present study, therefore, the ampli-tude of PhNR was measured from the peak of b-wave to thetrough level of PhNR.

Recent advances in OCT technology have revealed thepathomorphology of ME associated with BRVO, including thelocation of cystoid spaces and the usefulness of the junctionbetween the inner and outer segments of the foveal photore-ceptor layer as a hallmark of integrity of the outer retina.34–36

Previously, Tsujikawa et al.37 reported that ME in associationwith acute BRVO is accompanied frequently by serous retinaldetachment. However, it remains unclear whether serous ret-inal detachment causes the functional loss in BRVO. In thepresent study, foveal thickness showed a correlation with rel-ative amplitudes of the b-wave and PhNR. Furthermore, theserelative amplitudes were correlated with the height of serousretinal detachment but not with sensory retinal thickness.

TABLE 3. Comparison of Visual Acuity, Retinal Thickness, and fmERG at the Initial Visit and at 12Months in Each Group Stratified by the Treatments

No TreatmentGrid

Photocoagulation Vitrectomy

Eyes, n 12 8 4Sex, male/female 7/5 7/1 0/4Age, y 60.9 � 8.7 69.3 � 9.1 73.0 � 5.8Duration, mo 5.0 � 2.8 4.9 � 2.8 2.0 � 1.0Visual acuity, logMAR

Initial visit 0.52 � 0.47 0.58 � 0.22 0.44 � 0.3812 months 0.07 � 0.33* 0.39 � 0.27 0.63 � 0.51

Center point thickness, �mInitial visit 455 � 137 588 � 181 703 � 16212 months 270 � 99* 295 � 143* 446 � 236

a-Wave amplitude, �VInitial visit 0.9 � 0.4 1.0 � 0.5 0.9 � 0.212 months 1.2 � 0.5 0.8 � 0.6 0.7 � 0.4

b-Wave amplitude, �VInitial visit 1.7 � 0.6 1.5 � 0.8 1.5 � 0.912 months 2.4 � 0.7* 1.8 � 1.0 2.0 � 1.0

PhNR amplitude, �VInitial visit 2.4 � 1.1 1.8 � 1.0 1.8 � 0.912 months 3.1 � 1.2 2.2 � 1.3 2.4 � 1.0*

a-Wave latency, msInitial visit 21.5 � 1.5 24.2 � 3.3 23.9 � 3.312 months 23.2 � 3.3 22.6 � 2.3 20.9 � 1.4

b-Wave latency, msInitial visit 42.8 � 2.9 46.3 � 4.2 43.3 � 4.812 months 43.5 � 4.1 43.7 � 1.7 43.7 � 1.2

logMAR, logarithm of the minimum angle of resolution.* P � 0.05, compared with values at initial visit.

TABLE 4. Comparison of Parameters in fmERG between Ischemic and Nonischemic BRVO

Initial Examination Examination at 12 Months

Ischemic* Nonischemic P Ischemic* Nonischemic P

Relative amplitudea-Wave 0.64 � 0.30 0.88 � 0.51 0.120 0.78 � 0.32 0.59 � 0.24 0.140b-Wave 0.48 � 0.18 0.56 � 0.21 0.284 0.63 � 0.21 0.59 � 0.29 0.721PhNR 0.57 � 0.26 0.58 � 0.24 0.900 0.70 � 0.24 0.67 � 0.32 0.811

Relative latencya-Wave 1.05 � 0.08 1.10 � 0.17 0.329 1.06 � 0.12 1.08 � 0.13 0.699b-Wave 1.08 � 0.08 1.08 � 0.10 0.814 1.04 � 0.06 1.09 � 0.10 0.145

* Nonperfusion area of more than 5 disc diameters.



Therefore, we speculate that this correlation with retinal de-tachment might have contributed primarily to the correlationof center point thickness with relative amplitude. In addition,Table 2 indicates that macular function is negatively correlatedwith height of serous retinal detachment. As was shown inFigure 4, the extent of the serous retinal detachment may beassociated with impairment of function in the macular area. Ineyes with CRVO, it has been reported that the relative latencyhas a cross-correlation with sensory retinal thickness.26 In eyeswith ME associated with BRVO, however, the relative latencywas not correlated with any of the OCT parameters. This maybe explained partially by the fact that the extent of sensoryretinal swelling is milder in BRVO.

At 12 months after the initial visit, center point thicknesswas reduced significantly, with marked recovery of VA. OCTexamination at 12 months revealed that all affected eyesshowed substantial reduction in ME and complete absorbanceof the previously seen serous retinal detachment. In parallelwith the morphologic recovery of the macular area, fmERG at12 months also showed a recovery of some parameters; forexample, the amplitudes of the b-wave and of PhNR improvedsignificantly compared with those at the initial visit. The initialamplitude of the PhNR showed a correlation with the finalamplitude (r � 0.448; P � 0.035). Similar to VA, macularfunction after resolution of the ME associated with BRVOappears to be determined by the severity of impairment offunction during the acute phase. In the present study, persis-tent ME was treated with grid photocoagulation in 11 eyes andwith pars plana vitrectomy in four eyes. Table 3 shows param-eters of fmERG at the initial visit and at 12 months in eachgroup stratified by the treatment modality. Although thechanges of fmERG parameters showed a similar tendency ineach group, most were not statistically significant, perhapsbecause of the small number of eyes.

In addition, we assessed whether initial parameters offmERG served as prognostic factors of final VA. Unfortunately,however, we found no parameters of fmERG at the initial visitto be correlated with VA at 12 months (data not shown).

In the present study, initial fluorescein angiography showedan area of nonperfusion of �5 disc diameters in 44% of eyeswith BRVO. Between ischemic BRVO and nonischemic BRVO,there were no differences in fmERG parameters obtained at theinitial visit and at 12 months. Nonperfusion in the inner retinacould well cause functional impairment of the retina andwould have a substantial effect on the PhNR. Retinal ischemiacaused by BRVO, often seen outside the vascular arcade, didnot result in decreased fmERG, which reflected the only func-tion of the macular area.

Major limitations of the present study are its small samplesize and the various treatment regimens used. Although age,duration of disease, treatments, and other factors possiblyinfluenced the parameters of fmERG, small sample size did notallow multivariate regression analysis. However, this is the firstreport of fmERG used to evaluate macular function in eyes withME associated with BRVO and to compare the parameters offmERG with other measurements. We have demonstrated, withthe use of fmERG, that the amplitudes of each wave weresignificantly decreased in eyes with acute BRVO and showedimprovement after reductions in ME. Furthermore, impairedamplitudes of PhNR were correlated not with VA but, rather,with macular sensitivity, as measured with the MP1; thus, thePhNR in fmERG would effectively reflect the decreased macu-lar function from ME caused by BRVO. Recently, clinical trialsusing anti–vascular endothelial growth factor agents have beenperformed for the treatment of ME secondary to retinal veinocclusion.4,38,39 Such treatment, however, requires frequentintravitreal injection but does maintain a very low level ofvascular endothelial growth factor in the eye for a long time,

which might be a survival factor for the retinal neurons.40

fmERG, which allows for evaluation of macular function, maybe useful for monitoring the safety of these treatments.

References

1. The Branch Vein Occlusion Study Group. Argon laser photocoag-ulation for macular edema in branch vein occlusion. Am J Oph-thalmol. 1984;98:271–282.

2. Scott IU, Ip MS, Van Veldhuisen PC, et al. A randomized trialcomparing the efficacy and safety of intravitreal triamcinolonewith standard care to treat vision loss associated with macularedema secondary to branch retinal vein occlusion: the StandardCare vs Corticosteroid for Retinal Vein Occlusion (SCORE) studyreport 6. Arch Ophthalmol. 2009;127:1115–1128.

3. Gregori NZ, Rattan GH, Rosenfeld PJ, et al. Safety and efficacy ofintravitreal bevacizumab (Avastin) for the management of branchand hemiretinal vein occlusion. Retina. 2009;29:913–925.

4. Campochiaro PA, Heier JS, Feiner L, et al. Ranibizumab for macularedema following branch retinal vein occlusion: six-month primaryend point results of a phase III study. Ophthalmology. 2010;117:1102–1112.

5. Puliafito CA, Hee MR, Lin CP, et al. Imaging of macular diseaseswith optical coherence tomography. Ophthalmology. 1995;102:217–229.

6. Hee MR, Puliafito CA, Wong C, et al. Quantitative assessment ofmacular edema with optical coherence tomography. Arch Oph-thalmol. 1995;113:1019–1029.

7. Senturk F, Ozdemir H, Karacorlu M, Karacorlu SA, Uysal O. Micro-perimetric changes after intravitreal triamcinolone acetonide in-jection for macular edema due to central retinal vein occlusion.Retina. 2010;30:1254–1261.

8. Yamaike N, Kita M, Tsujikawa A, Miyamoto K, Yoshimura N.Perimetric sensitivity with the micro perimeter 1 and retinal thick-ness in patients with branch retinal vein occlusion. Am J Ophthal-mol. 2007;143:342–344.

9. Ogden TE, Hinton DR. Retina Basic Science and Inherited RetinalDisease. 3rd ed. Vol. 1. St. Louis: Mosby; 2001;317–331.

10. Chen H, Wu D, Huang S, Yan H. The photopic negative responseof the flash electroretinogram in retinal vein occlusion. Doc Oph-thalmol. 2006;113:53–59.

11. Hirose T, Miyake Y, Hara A. Simultaneous recording of electroreti-nogram and visual evoked response: focal stimulation under directobservation. Arch Ophthalmol. 1977;95:1205–1208.

12. Miyake Y, Ichikawa K, Shiose Y, Kawase Y. Hereditary maculardystrophy without visible fundus abnormality. Am J Ophthalmol.1989;108:292–299.

13. Terasaki H, Miyake Y, Tanikawa A, et al. Focal macular electroreti-nograms before and after successful macular hole surgery. Am JOphthalmol. 1998;125:204–213.

14. Terasaki H, Miyake Y, Nomura R, et al. Focal macular ERGs in eyesafter removal of macular ILM during macular hole surgery. InvestOphthalmol Vis Sci. 2001;42:229–234.

15. Hvarfner C, Andreasson S, Larsson J. Multifocal electroretinogra-phy and fluorescein angiography in retinal vein occlusion. Retina.2006;26:292–296.

16. Hvarfner C, Andreasson S, Larsson J. Multifocal electroretinogramin branch retinal vein occlusion. Am J Ophthalmol. 2003;136:1163–1165.

17. Chung EJ, Freeman WR, Koh HJ. Visual acuity and multifocalelectroretinographic changes after arteriovenous crossing sheatho-tomy for macular edema associated with branch retinal vein oc-clusion. Retina. 2008;28:220–225.

18. Seeliger M, Kretschmann U, Apfelstedt-Sylla E, Ruther K, ZrennerE. Multifocal electroretinography in retinitis pigmentosa. Am JOphthalmol. 1998;125:214–226.

19. Hood DC, Wladis EJ, Shady S, et al. Multifocal rod electroretino-grams. Invest Ophthalmol Vis Sci. 1998;39:1152–1162.

20. Fujii S, Escano MF, Ishibashi K, Matsuo H, Yamamoto M. Multifocalelectroretinography in patients with occult macular dystrophy.Br J Ophthalmol. 1999;83:879–880.



21. Tanikawa A, Horiguchi M, Kondo M, et al. Abnormal focal macularelectroretinograms in eyes with idiopathic epimacular membrane.Am J Ophthalmol. 1999;127:559–564.

22. Machida S, Gotoh Y, Toba Y, et al. Correlation between photopicnegative response and retinal nerve fiber layer thickness and opticdisc topography in glaucomatous eyes. Invest Ophthalmol Vis Sci.2008;49:2201–2207.

23. Machida S, Toba Y, Ohtaki A, et al. Photopic negative response offocal electroretinograms in glaucomatous eyes. Invest OphthalmolVis Sci. 2008;49:5636–5644.

24. Oishi A, Nakamura H, Tatsumi I, et al. Optical coherence tomo-graphic pattern and focal electroretinogram in patients with reti-nitis pigmentosa. Eye (Lond). 2009;23:299–303.

25. Nakamura H, Hangai M, Mori S, Hirose F, Yoshimura N. Hemi-spherical focal macular photopic negative response and macularinner retinal thickness in open-angle glaucoma. Am J Ophthalmol.2011;151:494–506.

26. Ogino K, Tsujikawa A, Nakamura H, et al. Focal macular electro-retinogram in macular edema secondary to central retinal veinocclusion. Invest Ophthalmol Vis Sci. 2011;52:3514–3520.

27. Viswanathan S, Frishman LJ, Robson JG, Harwerth RS, Smith EL3rd. The photopic negative response of the macaque electro-retinogram: reduction by experimental glaucoma. Invest Oph-thalmol Vis Sci. 1999;40:1124 –1136.

28. Kondo M, Kurimoto Y, Sakai T, et al. Recording focal macularphotopic negative response (PhNR) from monkeys. Invest Oph-thalmol Vis Sci. 2008;49:3544–3550.

29. McIntosh RL, Rogers SL, Lim L, et al. Natural history of centralretinal vein occlusion: an evidence-based systematic review. Oph-thalmology. 2010;117:1113–1123, e1115.

30. Rogers SL, McIntosh RL, Lim L, et al. Natural history of branchretinal vein occlusion: an evidence-based systematic review. Oph-thalmology. 2010;117:1094–1101, e1095.

31. Gundogan FC, Sobaci G, Bayraktar MZ. Intra-sessional and inter-sessional variability of multifocal electroretinogram. Doc Ophthal-mol. 2008;117:175–183.

32. Rangaswamy NV, Shirato S, Kaneko M, et al. Effects of spectral charac-teristics of Ganzfeld stimuli on the photopic negative response (PhNR) ofthe ERG. Invest Ophthalmol Vis Sci. 2007;48:4818–4828.

33. Viswanathan S, Frishman LJ, Robson JG. The uniform field and patternERG in macaques with experimental glaucoma: removal of spiking ac-tivity. Invest Ophthalmol Vis Sci. 2000;41:2797–2810.

34. Ota M, Tsujikawa A, Murakami T, et al. Association betweenintegrity of foveal photoreceptor layer and visual acuity in branchretinal vein occlusion. Br J Ophthalmol. 2007;91:1644–1649.

35. Ota M, Tsujikawa A, Kita M, et al. Integrity of foveal photore-ceptor layer in central retinal vein occlusion. Retina. 2008;28:1502–1508.

36. Ota M, Tsujikawa A, Murakami T, et al. Foveal photoreceptor layerin eyes with persistent cystoid macular edema associated withbranch retinal vein occlusion. Am J Ophthalmol. 2008;145:273–280.

37. Tsujikawa A, Sakamoto A, Ota M, et al. Serous retinal detachmentassociated with retinal vein occlusion. Am J Ophthalmol. 2010;149:291–301 e295.

38. Rosenfeld PJ, Fung AE, Puliafito CA. Optical coherence tomogra-phy findings after an intravitreal injection of bevacizumab (Avas-tin) for macular edema from central retinal vein occlusion. Oph-thalmic Surg Lasers Imaging. 2005;36:336–339.

39. Brown DM, Campochiaro PA, Singh RP, et al. Ranibizumab formacular edema following central retinal vein occlusion: six-monthprimary end point results of a phase III study. Ophthalmology.2010;117:1124–1133.

40. Nishijima K, Ng YS, Zhong L, et al. Vascular endothelial growthfactor-A is a survival factor for retinal neurons and a critical neu-roprotectant during the adaptive response to ischemic injury. Am JPathol. 2007;171:53–67.



Evaluation of Macular Function Using Focal Macular...

Documents

Transcript of Evaluation of Macular Function Using Focal Macular...