doi:10.1136/bjo.2007.135863 2008;92;1552-1557; originally published online 9 Jul 2008; Br. J. Ophthalmol.

  B L Sikorski, M Wojtkowski, J J Kaluzny, M Szkulmowski and A Kowalczyk  

dot syndromegreen angiography in multiple evanescent whitetomography with fluorescein and indocyanine Correlation of spectral optical coherence

http://bjo.bmj.com/cgi/content/full/92/11/1552Updated information and services can be found at:

These include:

References

  http://bjo.bmj.com/cgi/content/full/92/11/1552#BIBL

This article cites 20 articles, 3 of which can be accessed free at:

Rapid responses http://bjo.bmj.com/cgi/eletter-submit/92/11/1552

You can respond to this article at:

serviceEmail alerting

the top right corner of the article Receive free email alerts when new articles cite this article - sign up in the box at

Notes  

http://journals.bmj.com/cgi/reprintformTo order reprints of this article go to:

http://journals.bmj.com/subscriptions/ go to: British Journal of OphthalmologyTo subscribe to

on 27 October 2008 bjo.bmj.comDownloaded from

Correlation of spectral optical coherence tomographywith fluorescein and indocyanine green angiographyin multiple evanescent white dot syndrome

B L Sikorski,1 M Wojtkowski,2 J J Kaluzny,1 M Szkulmowski,2 A Kowalczyk2

1 Department of Ophthalmology,Collegium Medicum, NicolausCopernicus University,Bydgoszcz, Poland; 2 Institute ofPhysics, Nicolaus CopernicusUniversity, Torun, Poland

Correspondence to:Dr B L Sikorski, Department ofOphthalmology, NicolausCopernicus University, 9 M.Sklodowskiej-Curie Street, PL85-094, Bydgoszcz, Poland;sikorski@doctors.org.uk

Accepted 28 May 2008Published Online First9 July 2008

ABSTRACTAims: To determine the spatial location of lesions inmultiple evanescent white dot syndrome (MEWDS) withthe aid of spectral optical coherence topography (SOCT),fluorescein angiography (FA) and indocyanine greenangiography (ICGA).Methods: A novel method of three-dimensional SOCTdata analysis called reflectivity maps was introduced. Thereflectivity maps display the distribution of a back-reflected intensity taken only from individual retinal layerslocated at specific distance from the reference plane.Reflectivity maps of the inner retina, the junction betweenphotoreceptor inner and outer segments (IS/OS), retinalpigment epithelium and choroid of the patient withMEWDS were created and correlated with FA and ICGA.Results: During the acute stage of MEWDS, thereflectivity map of the IS/OS junction displayed areas ofreduced reflectivity that showed a strong positivecorrelation with hypofluorescent ICGA spots and a weakerbut positive correlation with hyperfluorescent FA dots.SOCT examination did not reveal any pathological changesinvolving either any other retinal layers or the innerchoroid.Conclusion: Disseminated disruptions of the IS/OSjunction seen on SOCT cross-sectional images in theacute stage of MEWDS form the pattern of spots that canbe correlated with those revealed by ICGA. This suggeststhat hypofluorescent ICGA spots indicate alternations inthe retinal pigment epithelium–photoreceptor complexand do not represent inflammatory choroidal lesions.

Multiple evanescent white dot syndrome(MEWDS), first described in 1984 by Jampol et al,belongs to a group of rare multifocal inflammatoryconditions involving the retina.1 Patients sufferingfrom this syndrome experience acute, painless,unilateral loss of vision. They may also reportphotopsia, dyschromatopsia and temporal or para-central scotomas. A characteristic fundus finding isthe presence of multiple small, ill-defined whitedots scattered over the posterior pole and mid-periphery, located at the level of the outer retina orretinal pigment epithelium (RPE).1 Other clinicalfindings may include a granular appearance of themacula, retinal vascular sheathing, vitreous cellsand optic disc oedema. Fluorescein angiography(FA) demonstrates early and late hyperfluorescenceof the white dots. Intermediate and late-phaseindocyanine green angiography (ICGA) revealsmultiple small hypofluorescent spots that are morenumerous than those seen both clinically and withFA.2 Electrophysiological studies demonstrate areduced a-wave on the ERG.3 Visual field test

reveals an enlarged blind spot. MEWDS has notbeen associated with any systemic conditions.However, a preceding viral illness has beenreported in up to 50% of patients.1 The disease isself-limiting with almost all patients regaininggood visual acuity within several weeks.Recurrence is unusual but has been reported.4

The pathogenesis of MEWDS is still unknown,but there is a probability of an infectious orautoimmune aetiology of the disease.5 6

Controversy also exists as to whether the initiallesions in this disease are located only in the outerretina or in the retina and choroid.1 2 3 7 8

Kanis and van Norren reported in a 20-year-oldfemale patient with MEWDS that the junctionbetween photoreceptor inner and outer segments(IS/OS) in the foveal region, during the acute phaseof the disease, was not visible in optical coherencetopography (OCT) cross-sectional images.9 After21 weeks, OCT examination revealed the IS/OSjunction. Nguyen et al presented similar findings inultrahigh-resolution OCT in the acute stage butdid not perform imagining after recovery.10

However, to the best of our knowledge, there havebeen no reports showing either the character or thedistribution of observed disturbances, or correlat-ing them with FA and ICGA.

In this paper, we applied recently developedspecialised spectral optical coherence topography(SOCT) measurement protocols and custom-designed software to analyse retinal structure of apatient with MEWDS.11 Using three-dimensionaldata, it was possible to extract the maps displayingreflectivity of the individual SOCT layers includingIS/OS junction and RPE. We also analysed thereflectivity of the inner retina and inner choroid.The reflectivity maps were compared with FA andICGA.

METHODSTomographic data presented in this paper wereobtained by the prototype high-resolution, high-speed SOCT system developed and constructed atNicolaus Copernicus University, Poland.11 12 TheSOCT instrument provides three-dimensionalimages with 4.5 mm axial and 15 mm transverseresolution, and an acquisition speed of 25 000 axiallines per second. The device uses Broadlighter D830light source (Superlum, Moscow) emitting light ofcentral wavelength 830 nm and 70 nm of fullwidth at half maximum.

In order to unveil subtle disturbances in theretina, two measurement protocols have beenused. To collect general information on structural

Innovations

1552 Br J Ophthalmol 2008;92:1552–1557. doi:10.1136/bjo.2007.135863

on 27 October 2008 bjo.bmj.comDownloaded from

changes, 35 cross-sectional images, each consisting of 3000 A-scans, covering the area of 663 mm, were taken. Anotherprotocol provided three-dimensional data to create reflectivitymaps of individual SOCT layers. The latter one measured 200cross-sectional images, each consisting of 400 A-scans, coveringthe area of 565 mm. The scanning was performed in a rasterpattern which enabled uniform reconstruction of the fundusimage.

The data were arranged and displayed as classical 2D cross-sectional images (optical B-scans) and reflectivity maps ofSOCT layers, a method introduced in this manuscript. Thereconstruction of reflectivity maps of the RPE, inner choroid, IS/OS and the inner retina is performed in the following steps.First, the outline of the complex of the IS/OS and RPE at all 200cross-sectional images is automatically marked using custom-designed software, which has been described elsewhere(fig 1B,D).11 The posterior part of the delineated contour isthen used to fit the smooth parabolic curve called the outerretinal contour (ORC). The ORC is then selected as a referenceline dividing tomograms into two parts: the upper onecontaining the retina and posterior vitreous and the lower onecomprising the choroid. In the next step, each 2D cross-sectionalimage is realigned to flatten the ORC while preserving theoriginal radial dimensions of the retina. Once the ORC istransformed into a straight line, it is possible to createreflectivity maps displaying the distribution of a back-reflectedintensity from individual retinal layers located at specificdistance from the reference plane consisting of the flatORCs.13 14 In particular, we manually choose a region in theflattened cross-sectional image that contains the layer ofinterest. The entire backscattered intensity from the chosenregion is added in axial direction providing a single line of thereflectivity map. The lines obtained from all cross-sectionalimages form the reflectivity map of the layer. A special type ofreflectivity maps are fundus view maps, which display the totalsignal returning from the entire depth of the imaged retina(fig 1A,C). The fundus view map is analogous to the image fromthe scanning laser ophthalmoscope.14 15

The maps reconstructed from SOCT-derived spatial datawere digitally overlaid onto the fundus images obtained fromFA (TRC 50 XE, Topcon, Tokyo, Japan) and ICGA (Angioscan-300 LDT, Mount Vernon, NY) by matching patterns of retinalvasculature.

The patient was examined twice: once during the acute phaseof the disease and then after recovery. The optical powerincident on the eye was 750 mW. The examination wasperformed under the tenants of the Helsinki Agreement, theprotocol was approved by the University Ethics Commissionand a written consent was obtained from the patient and hisparents.

CASE REPORTA 15-year-old boy presented with a complaint of decreasedvision in his right eye for 3 days. He reported having severe flu-like symptoms 3 weeks before visiting our hospital. Thosesymptoms disappeared within a week. On admission, the visualacuity was 20/40 in the right eye and 20/20 in the left. Theintraocular pressure was 12 and 11 mm Hg in the right and lefteye, respectively. The anterior segments were unremarkable.The dilated fundus examination showed no evidence of vitreousinflammatory activity in both eyes. Optic nerve heads lookednormal. In the right eye, however, the fovea appeared granular,and several small faint white spots were visible throughout theposterior pole. Four weeks after admission, the fundus lesionswere no longer visible, and the vision returned to normal.

RESULTSDuring the acute stage of MEWDS, FA of the affected eyerevealed early wreath-like hyperfluorescence (fig 2A) with latestaining (fig 2B). However, fundus colour photographyperformed at the same time demonstrated only few barelyvisible white lesions. An intermediate and late-phase ICGAshowed numerous hypofluorescent spots in the posterior poleand mid-periphery that outnumbered those seen on FA (fig 3A).The spots appeared approximately 9 min after ICG dyeinjection. SOCT cross-sectional images of the macular regionrevealed a strong disruption of the line corresponding to the IS/OS junction by randomly distributed regions of weaker lightbackscattering intensity (fig 4A). The reflectivity map for theIS/OS junction showed the areas of strong signal attenuation(darker regions), while the reflectivity map for the inner retinadid not reveal any abnormalities (fig 5B,C). There were nohypo- or hyper-reflective areas noticeable in the reflectivitymaps of the RPE and inner choroid (fig 5D,E). Comparison ofhyporeflective SOCT areas derived from the reflectivity map of

Figure 1 Three-dimensional, high-resolution spectral optical coherencetopography (SOCT) imaging of the retinaof a patient suffering from multipleevanescent white dot syndrome. (A, B)Data collected during the acute phase ofthe disease. (C, D) Data collected afterrecovery. (A, C) SOCT fundus viewsreconstructed from three-dimensionaldata measured in raster protocol coveringarea of 565 mm. (B, D) Exemplary cross-sectional SOCT images from the set of200 optical B-scans with a superimposedred contour indicating the automaticallysegmented band of the inner/outersegments junction retinal pigmentepithelium complex. A bright ring in (C) isdue to a specular reflection from theretina.

Innovations

Br J Ophthalmol 2008;92:1552–1557. doi:10.1136/bjo.2007.135863 1553

on 27 October 2008 bjo.bmj.comDownloaded from

the IS/OS junction with hypofluorescent ICGA spots revealed ahigh degree of shape, structure, distribution and size resemblance(fig 3B). However, not all lesions visible on the IS/OS junctionreflectivity map were represented in the fluorescein image, andsome were slightly larger than those seen on FA (fig 2C).

Four weeks after presentation to the hospital, visual acuityrecovered to 20/20. Clinical examination, FA and ICGA showedresolution of the fundus lesions. Some residual RPE granularitypersisted in the fovea region. On the SOCT image, a linecorresponding to the IS/OS junction became continuous (fig 4B).Reflectivity map of the IS/OS junction displayed no areas of

Figure 2 (A) Arteriovenous phase fluorescein angiogram of theaffected eye during the acute phase of multiple evanescent white dotsyndrome showing mild hyperfluorescence of the retinal lesions. Notethe punctate hyperfluorescent specks in the centre of the macula smallerthan the white dots seen elsewhere. (B) Fluorescein angiogram 5 minafter dye injection (acute stage) indicating an increase in thehyperfluorescent areas due to staining (orange arrows). (C) Reflectivitymap of the inner/outer segments (IS/OS) junction superimposed ontofluorescein angiography (FA). Note the high degree of internal structureresemblance of the spot indicated by the black arrow and revealedindependently by fundus FA and spectral optical coherence topography

Figure 3 (A) Late-phase ICG angiogram of the affected eye, 26 min afterdye injection during the acute phase of multiple evanescent white dotsyndrome, showing hypofluorescent spots in the posterior pole. (B)Reflectivity map of the inner/outer segments (IS/OS) junction superimposedonto the indocyanine green angiogram. Note the high degree of internalstructure resemblance of the spot pointed by the black arrow and revealedindependently by the fundus indocyanine green angiography (ICGA) andspectral optical coherence topography reflectivity map. Even tinydisturbances seen on ICGA are present in the reflectivity map of the IS/OSjunction in a pronounced manner (white arrow).

(SOCT) reflectivity map. Even several tiny disturbances seen on FA arealso expressed in the reflectivity map of the IS/OS junction (whitearrow). However, not all SOCT lesions are present in the fluoresceinpicture (red arrow). Also, some lesions visible on the reflectivity map arelarger than those seen on FA (green arrow).

Innovations

1554 Br J Ophthalmol 2008;92:1552–1557. doi:10.1136/bjo.2007.135863

on 27 October 2008 bjo.bmj.comDownloaded from

signal attenuation. There was no evidence of pathologicallesions involving either any other retinal layers or the innerchoroid at any stage of the disease.

DISCUSSIONSince the description of MEWDS, there has been a raging andcontinuing controversy over the exact depth location of funduslesions in this disease. Some authors suggest that the evanescentlesions are situated in the outer retina or RPE, while othersbelieve that hypofluorescent spots on ICGA are the results ofchoroidal abnormalities.1 2 3 These hypotheses are based onclinical presentation, angiographical appearance of lesions andelectrophysiological evidence.1 2 3 7 8 16 In this paper, we used forthe first time the 3D SOCT technique to obtain additionalinformation on this disease. In the acute stage, we found stronginhomogeneity on the reflectivity of a surface corresponding tothe IS/OS junction. This might indicate the degeneration ormetabolic disturbance of photoreceptors. However, it seemsvery unlikely that a degenerative process could be fullyreversible within a few weeks. We suppose that areas of lowerreflectivity of the IS/OS junction can represent inflammatoryswelling of the photoreceptors. A reconstructed reflectivity mapof this layer showed that disseminated disruptions of the IS/OSjunction noticeable on SOCT cross-sectional images are notarranged in a random manner. The regions of the reducedreflectivity of the IS/OS junction form the pattern of spots. Thespatial arrangement of this pattern as well as the shape, size andinternal structure of particular spots is strikingly similar to thatof hypofluorescent ICGA dots. Since there is no direct linearanatomical connection between the retina and the innerchoroid, as well as inside the retina alone, it is very unlikelythat these disturbances, visualised independently by both

techniques, result from two different types of lesions locatedin the exact same vertical plane but at various depth levels ofthe retina and/or the choroid. Also, reflectivity maps of theinner retina, RPE and inner choroid did not show any abnormalchanges in intensity of backscatter light. This strongly indicatesthat the hypofluorescent spots seen on ICGA reflect alterna-tions in the photoreceptors or in the RPE and photoreceptors, asthey form a functional complex.17 The question arises as to howthe disturbances in these structures can cause intermediate andlate hypofluorescence on ICGA. Chang et al reported that theICG fluorescence detected on histological sections of monkeyand human eyes was seen in the extravascular choroidal stromawithin 10 min after dye injection.18 They also found that theICG accumulated within the intact RPE and proposed that thisaccumulation contributed to the background hyperfluorescenceon clinical ICGA. Chang et al also reported that in a patient whounderwent the surgical excision of choroidal neovascularisation15 min after the ICG dye injection, the intact RPE removedtogether with the membrane was intensely fluorescent oninfrared fluorescence microscopy.19 They found the fluorescenceof the RPE to be even more intense than that of fibrovasculartissue. This indicates that RPE cells accumulate ICG. Wepresume that in MEWDS, a dysfunctional RPE–photoreceptorcomplex might not accumulate the ICG dye in the intermediateand late phase of angiography. This could account for hypofluor-escent lesions seen on ICGA. The results of SOCT examinationalso support the assumption that hypofluorescent spots on ICGAare not caused by blockage of background choroidal fluorescence.The central wavelength of the electromagnetic radiation used inour SOCT device and ICGA are from the same spectral range. Aswe observed no blockage of backscattering light reflected from theRPE and choroid on SOCT (fig 5D,E), it can be expected that thereis also no blockage of light recorded from the choroid by ICGAinstruments.

Gross et al speculated that hypofluorescent ICGA spotscould represent the failure of the ICG molecules to conjugatewith possible shallow inflammatory infiltration within theinner choroid, leaving staining of the extrachoroidal vascularspace everywhere except at the spot sites.7 Since we were notable to see any choroidal changes in the SOCT maps and cross-sectional images, we believe that the mechanism described byGross et al is not responsible for the formation of hypo- andhyperfluorescent dots visible in ICGA and FA in the casedescribed.

We have also investigated the degree of resemblance betweenthe SOCT reflectivity map of the IS/OS junction and FA image.We found that some lesions visible on the IS/OS junctionreflectivity map were slightly larger than those seen on FA andthat not all lesions revealed by SOCT were represented on thefluorescein picture. As we believe that SOCT, ICGA and FAdepict the same lesions, the question arises as to why most areasof the photoreceptors disturbances visualised by SOCT reflec-tivity map are reflected in FA but some are not.

We suggest the following explanation of this phenomenon.The early hyperfluorescent spots on FA are seen only in regionsof alteration in the outer blood–retina barrier. Fluoresceinmolecules may then infiltrate from extravascular choroidalspace through dysfunctional RPE. However, not all disturbancesvisualised by SOCT and ICGA or even not all within one singlespot-shaped area are connected with breakdown of the barrier.That is why more spots are seen on SOCT and ICGA than onFA. The results of scanning laser densitometry, in the case ofMEWDS, revealed that several lesions with no visual pigmentwere slightly larger than the hyperfluorescent spots in FA

Figure 4 High-quality spectral optical coherence topography cross-sectional retinal images of a patient suffering from multiple evanescentwhite dot syndrome. (A) Data collected during acute phase of thedisease revealing strong inhomogeneity on the reflectivity of a linecorresponding to the inner/outer segments junction (white arrow). (B)Data collected after regaining visual acuity to 20/20. Note the normalappearance of the retina. Scale bars correspond to 100 mm.

Innovations

Br J Ophthalmol 2008;92:1552–1557. doi:10.1136/bjo.2007.135863 1555

on 27 October 2008 bjo.bmj.comDownloaded from

examination.20 This finding seems to be consistent with theexplanation we propose.

The results of SOCT examination that we presented herein arealso compatible with electrophysiological studies in MEWDS thatpoint to transient disturbance at the level of the RPE andphotoreceptor outer segments complex. Sieving et al demonstratedimpaired photoreceptor function (abnormal a-wave), markedlyreduced visual pigment optical density of the outer segments (earlyreceptor potential (ERP) amplitude) and abnormal visual pigmentregeneration (ERP regeneration kinetics) during the acute stage ofMEWDS.3 Feigl et al found supernormal N1- and P1-waveamplitudes of the mf-ERG first-order kernel in the early courseof the disease.16

Our results suggest that hypofluorescent spots seen on ICGAreflect alternations in photoreceptors or dysfunction in theRPE–photoreceptor complex. By any interpretation, these spotsdo not represent lesions located in the choroid. Obviously, thisdoes not imply that MEWDS cannot have choroidal compo-nents. Since the pathogenetic factors of the disease are stillunknown, it cannot be ruled out that these factors can alsoinduce changes in the retinal circulation, RPE or inner choroidwhich are beyond the imagining detection of all three modalitieswhich we used. Further detailed studies are therefore required toinvestigate the precise pathophysiology of MEWDS. Futurework is also needed to better understand the interpretation ofouter retina disturbances in SOCT examination.

Competing interests: None.

Ethics approval: Ethics approval was provided by the University Ethics Commission.

Patient consent: Parental consent was obtained.

REFERENCES1. Jampol LM, Sieving PA, Pugh D, et al. Multiple evanescent white dot syndrome. I.

Clinical findings. Arch Ophthalmol 1984;102:671–4.2. Ie D, Glaser BM, Murphy RP, et al. Indocyanine green angiography in multiple

evanescent white dot syndrome. Am J Ophthalmol 1994;117:7–12.3. Sieving PA, Fishman GA, Jampol LM, et al. Multiple evanescent white dot

syndrome. II Electrophysiology of the photoreceptors during retinal pigment epithelialdisease. Arch Ophthalmol 1984;102:675–9.

4. Tsai L, Jampol LM, Pollock SC, et al. Chronic recurrent multiple evanescent white dotsyndrome. Retina 1994;14:160–3.

5. Jampol LM, Becker KG. White spot syndromes of the retina: a hypothesis based onthe common genetic hypothesis of autoimmune/inflammatory disease.Am J Ophthalmol 2003;135:376–9.

6. Gass JD. Are acute zonal occult outer retinopathy and the white spot syndromes(AZOOR complex) specific autoimmune diseases? Am J Ophthalmol 2003;135:380–1.

7. Gross NE, Yannuzzi LA, Freund KB, et al. Multiple evanescent white dot syndrome.Arch Ophthalmol 2006;124:493–500.

8. Obana A, Kusumi M, Miki T. Indocyanine green angiographic aspects of multipleevanescent white dot syndrome. Retina 1996;16:97–104.

9. Kanis MJ, van Norren D. Integrity of foveal cones in multiple evanescent white dotsyndrome assessed with OCT and foveal reflection analyser. Br J Ophthalmol2006;90:795–6.

10. Nguyen MH, Witkin AJ, Reichel E, et al. Microstructural Abnormalities in MEWDSDemonstrated by Ultrahigh Resolution Optical Coherence Tomography. Retina2007;27:414–18.

11. Szkulmowski M, Wojtkowski M, Sikorski B, et al. Analysis of posterior retinal layersin spectral optical coherence tomography images of the normal retina and retinalpathologies. J Biomed Opt 2007;12:041207.

12. Wojtkowski M, Leigeb R, Kowalczyk A, et al. In vivo human retinal imaging byFourier domain optical coherence tomography. J Biomed Opt 2002;7:457–63.

Figure 5 Spectral optical coherence topography reflectivity maps in the acute stage of multiple evanescent white dot syndrome. (A) Fundus view forthe entire retina; the bright areas are specular reflections from the surface of the retina, while the dark streaks are shadows cast by retinal bloodvessels. (B) Reflectivity map of the inner retina showing no alteration in intensity of backscatter light; reflections from retinal vessels can be seen. (C)Reflectivity map of the inner/outer segments junction; the dark spots correspond to the areas of weaker light backscattering. (D) Reflectivity map of theretinal pigment epithelium layer demonstrating no abnormal changes in light-backscattering intensity. (E) Reflectivity map of the inner choroid revealingno alternation in the amount of light reflected back from this layer. The imaging region in all maps is 565 mm.

Innovations

1556 Br J Ophthalmol 2008;92:1552–1557. doi:10.1136/bjo.2007.135863

on 27 October 2008 bjo.bmj.comDownloaded from

13. Jiao SL, Knighton R, Huang XR, et al. Simultaneous acquisition of sectional andfundus ophthalmic images with spectral-domain optical coherence tomography.Optics Express 2005;13:444–52.

14. Wojtkowski M, Srinivasan V, Fujimoto JG, et al. Three-dimensional retinal imagingwith high-speed ultrahigh-resolution optical coherence tomography. Ophthalmology2005;112:1734–46.

15. Wojtkowski M, Bajraszewski T, Gorczynska I, et al. Ophthalmic imaging by spectraloptical coherence tomography. Am J Ophthalmol 2004;138:412–19.

16. Feigl B, Haas A, El-Shabrawi Y. Multifocal ERG in multiple evanescent white dotsyndrome. Graefes Arch Clin Exp Ophthalmol 2002;240:615–21.

17. Steinberg RH. Interactions between the retinal pigment epithelium and the neuralretina. Doc Ophthalmol 1985;60:327–46.

18. Chang AA, Morse LS, Handa JT, et al. Histologic localization of indocyanine greendye in aging primate and human ocular tissues with clinical angiographic correlation.Ophthalmology 1998;105:1060–8.

19. Chang AA, Zhu M, Billson FA, et al. Indocyanine green localisation in surgicallyexcised choroidal neovascular membrane in age related macular degeneration.Br J Ophthalmol 2004;88:307–9.

20. van Meel GJ, Keunen JE, van Norren D, et al. Scanning laser densitometry inmultiple evanescent white dot syndrome. Retina 1993;13:29–35.

Access all our original articles online even before they appear in a print issue!

Online First is an exciting innovation that allows the latest clinical research papers to go fromacceptance to your browser within days, keeping you at the cutting edge of medicine.

Simply follow the Online First link on the homepage and read the latest Online First articles that areavailable as unedited manuscripts in downloadable PDF form. The articles are peer reviewed, acceptedfor publication and indexed by PubMed but not yet included in a journal issue, so you’ll be among thefirst to read them!

Innovations

Br J Ophthalmol 2008;92:1552–1557. doi:10.1136/bjo.2007.135863 1557

on 27 October 2008 bjo.bmj.comDownloaded from