SPECIAL ARTICLE Mu¨ ller Cell Cone, an Overlooked Part of the Anatomy of the Hypotheses Concerning Its Role in the Pathogenesis of Macular Hole and Foveomacular Retinoschisis

J. Donald M. Gass, MD

oorly recognized by anatomists and pathologists is the cone-shaped zone of Mu¨ ller cells that composes the central and inner part of the fovea centralis. The importance of these cells in the structural integrity of the macula, as a repository for , and in the pathogenesis of macular diseases, particularly regarding idiopathic macular hole and foveomacularP schisis, is hypothesized. Arch Ophthalmol. 1999;117:821-823 Some standard textbooks of ocular outer limiting membrane, where there anatomy and pathology have fostered the were no receptor nuclei (arrowhead, misconception that the is com- Figure). The base of the cone formed the posed solely of retinal cone cells.1,2 Largely floor of the fovea centralis and extended overlooked has been the report of Ya- into the area of the in the perifo- mada,3 who investigated the electron mi- veolar region (arrows, Figure). In the light croscopic findings of the fovea centralis. microscopic thin sections prepared for He found an inverted cone-shaped zone electron microscopy, the cytoplasm of the of specialized Mu¨ ller cells composing the Mu¨ ller cells composing the outer portion floor of the fovea (Mu¨ ller cell cone). of the Mu¨ ller cell cone appeared opti- cally empty. The cytoplasm became slightly See also page 744 more dense toward the base of the Mu¨ ller cell cone. The low density of the Mu¨ ller The purposes of this report are to re- cell cytoplasm within the cone was evi- view the ultrastructure of the fovea cen- dent in electron micrographs and was in tralis and to present the hypotheses that contrast to the greater density of Mu¨ ller the Mu¨ ller cell cone is (1) a reservoir for cell cytoplasm elsewhere in the concentrated retinal xanthophyll, (2) the where the Mu¨ ller cells were intimately in- primary structural support for the fo- tertwined with the receptor neurites. The veola, (3) a primary role player in age- Mu¨ ller cell cone was free of neurites ex- related macular hole formation, and (4) the cept near its apex, where the long outer primary locus of changes causing the pa- cone fibers extended from the outer lim- thognomonic biomicroscopic signs of fo- iting membrane anteriorly and out- veomacular schisis. wardly in a radiating fashion through Mu¨ ller cell cytoplasm into the surround- ULTRASTRUCTURE OF THE ing cone nuclei (arrowhead, Figure). A few FOVEA CENTRALIS nuclei of either atypical glial cells or ec- topic ganglion cells were present within In 1969, Yamada3 reported the light and the Mu¨ ller cell cone. Not stated in Yama- electron microscopic anatomy of the fo- da’s report was the location of the nuclei vea centralis in the of a 45-year-old of the Mu¨ ller cells composing the cone. woman. The inner half of the foveola was The internal limiting membrane lining the composed of an inverted cone-shaped zone of Mu¨ ller cells (Mu¨ ller cell cone). The trun- cated apex of the cone was located at the This article is also available on our Web site: www.ama-assn.org/ophth. From the Department of Ophthalmology, Vanderbilt University, Nashville, Tenn.

ARCH OPHTHALMOL / VOL 117, JUNE 1999 821

©1999 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 contraction of the prefoveolar vit- reous cortex that is tightly adher- ent to the internal limiting mem- g brane of the Mu¨ ller cell cone. It is Mcc probable that Mu¨ ller cell invasion and proliferation within the prefo- H veolar vitreous cortex are impor- tant in causing contraction of the prefoveolar vitreous cortex and the sequence of events postulated above. Recently, investigators using opti- Drawing of the anatomy of the fovea centralis showing Mu¨ller cell cone (Mcc) whose base (arrows) cal coherence tomography and scan- corresponds with the internal limiting membrane, and whose apex corresponds with the outer limiting ning electron microscopy to study membrane centrally (arrowhead). Henle fiber layer (H) and foveal edge of the (g) patients with impending macular are shown. Drawing represents the author’s interpretation of a photomicrograph from Yamada.3 holes have demonstrated evidence of a split in the foveola with cyst for- inner surface of the Mu¨ ller cell cone gion, xanthophyll is largely con- mation in some patients.8,9 The was thin (10 nm to 20 nm) com- fined to the outer nuclear and plexi- Mu¨ ller cell cone provides an ana- pared with that in the peripheral fo- form layers throughout the macula. tomical substrate for schisis to oc- veal area (1.5 µm). Hogan and co- Xanthophyll becomes progressively cur. A split occurring within the reti- workers4 subsequently confirmed less concentrated toward the periph- nal receptor cell layer would be Yamada’s findings. eral macula. The presence of highly unlikely. Madreperla et al10 and Ezra concentrated xanthophyll in pseudo- et al7 studied 20 opercula excised HYPOTHESES opercula in patients with various during surgery for stage 3 macular stages of age-related macular holes is holes. They found native vitreous Hypothesis 1: Mu¨ ller cell cone is a additional evidence in support of the cortex, as well as Mu¨ ller cells and/or reservoir for retinal xanthophyll. presence of concentrated xantho- fibrous astrocytes, in 100% of cases. Xanthophyll is concentrated within phyll in the Mu¨ ller cell cone.6,7 In approximately 50% they found the cells that make up the Mu¨ ller cell Hypothesis 2: The Mu¨ ller cell bits of the inner retina including in- cone and is partly responsible for the cone is the primary structural sup- ternal limiting membrane, neu- peculiar low density of the cell cy- port for the fovea. The Mu¨ ller cell rites, and some perikarya. This is evi- toplasm in this area. cone serves as a plug to bind to- dence suggesting that contraction of Investigators have found that gether the receptor cells in the fo- the prefoveolar vitreous cortex in identifying the exact location of the veola. Without this plug of glial cells, some cases causes avulsion of the in- retinal xanthophyll is difficult. Snod- the retinal receptor cell layer with its ternal limiting membrane, the Mu¨ ller derly and coworkers5 reported that thin layer of horizontally radiating cell cone, superficial inner cone fi- the retinal xanthophyll is most con- nerve fibers would be highly sus- bers (Henle layer), and occasion- centrated within the inner part of the ceptible to disruption at the umbo ally a few cone nuclei. Failure to find foveola and perifoveolar area in what and hole formation under a variety inner retinal elements in 50% of was thought to be the fiber layers (re- of circumstances, including sen- cases suggests that contraction of the ceptor axon layer and inner plexi- sory retinal detachment, minor prefoveolar vitreous cortex in the ab- form layer). Since there is minimal trauma, cystoid macular edema, and sence of avulsion of the inner retina nerve fiber layer in the foveolar area, . is capable of causing dissolution of it is probable that most of the xan- Hypothesis 3: The Mu¨ ller cell the Mu¨ ller cell cone and dehis- thophyll there is present within the cone is a primary role player in cence of the foveola at the umbo. Mu¨ ller cells and accounts for the low age-related macular hole forma- Failure to find large numbers of cone density of their cytoplasm within the tion. The earliest changes occur- nuclei, outer limiting membrane el- area of the cone. Likewise, most of the ring in the development of age- ements, and inner cone receptors is superficial layer of perifoveolar xan- related macular hole involve evidence that horseshoe-shaped fo- thophyll may reside within the cells degeneration of the Mu¨ ller cell cone– veolar tears and formation of full- that compose the base of the Mu¨ ller vitreous cortex interface, Mu¨ ller cell thickness retinal opercula occur cell cone rather than in the inner invasion and proliferation within the rarely in age-related macular hole plexiform layer. The superficial reti- prefoveolar vitreous cortex, contrac- formation. Separation of the vitre- nal xanthophyll, which extends only tion of the prefoveolar vitreous cor- ous cortex in the perifoveolar area a short distance beyond the foveal tex, disruption of the Mu¨ ller cell is evident by ultrasound and opti- capillary-free zone into the region of cone, full-thickness retinal dehis- cal coherence tomography in the the clivus, is probably responsible for cence at the umbo, centrifugal re- normal fellow of some pa- the superficial retinal burns that oc- traction of the receptor cell layer, and tients with macular holes and in cur as applications of argon blue- formation of a prehole opacity. many patients with impending green photocoagulation are placed There is considerable evi- macular holes.9,11 Some authors have near the edge of the capillary-free dence that idiopathic age-related suggested that tangential traction ex- zone. Beyond the perifoveolar re- macular hole formation begins with erted on the fovea by the normal

ARCH OPHTHALMOL / VOL 117, JUNE 1999 822

©1999 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 movement of the may schisis at the time of enucleation. Textbook. 3rd ed. Philadelphia, Pa: WB Saun- be responsible for macular hole de- Before Yamada’s ultrastructural ders Co; 1985;2:598. 12 3. Yamada E. Some structural features of the fovea velopment. It is not known, how- study of the normal fovea, if one centralis in the human retina. Arch Ophthalmol. ever, whether perifoveal separation considered that there were no 1969;82:151-159. of the vitreous is the cause of, the re- Mu¨ ller cells or ganglion cells and 4. Hogan MJ, Alvarado JA, Weddell JE. Histology of sult of, or unrelated to contraction only a few nerve fibers in the the : An Atlas and Textbook. Philadel- phia, Pa: WB Saunders Co; 1971:492-497. of the prefoveolar cortex. foveola, there was no plausible 5. Snodderly DM, Auran JD, Delori FC. The macular Hypothesis 4: Congenital explanation for foveomacular pigment: spacial distribution in . abnormalities affecting the Mu¨ ller schisis. Invest Ophthalmol Vis Sci. 1984;25:674-685. cell cone are responsible for the Further investigation of the ul- 6. Gass JDM, Van Newkirk M. Xanthic scotoma and pathognomonic biomicroscopic trastructure of the fovea centralis in yellow foveolar shadow caused by pseudo- operculum after vitreoretinal separation. Retina. picture of foveomacular patients of different ages and sex are 1992;12:242-244. schisis. X-linked juvenile retinos- needed to give us more insight con- 7. Ezra E, Munro PMG, Charteris DG, et al. Macular hole chisis is characterized clinically by cerning the pathogenesis of age- opercula. Arch Ophthalmol. 1997;115:1381-1387. peculiar macular changes (foveo- related macular hole and other dis- 8. Kishi S, Kamei Y, Shimizu K. Tractional elevation of Henle’s fiber layer in idiopathic macular holes. macular schisis) in 100% of chil- orders affecting the macula. Am J Ophthalmol. 1995;120:486-496. dren and young adults and in 9. Hee MR, Puliafito CA, Wong C. Optical coher- approximately 50% of patients by Accepted for publication November 6, ence tomography of macular holes. Ophthalmol- peripheral retinoschisis and 1998. ogy. 1995;102:748-756. so-called vitreous veils. Later in life 10. Madreperla SA, McCuen II Jr, Hickingbotham D, This study was supported in part Green WR. Clinicopathologic correlation of sur- the delicate superficial radiating by a Senior Scientific Award from Re- gically removed macular hole opercula. Am J Oph- network of fine polycystic changes, search to Prevent Blindness Inc, New thalmol. 1995;120:197-207. characteristic of foveomacular schi- York, NY. 11. Van Newkirk MR, Gass JDM, Callanan D, et al. Fol- sis, may be replaced by nonspecific Reprints: J. Donald M. Gass, MD, low-up and ultrasonographic examination of pa- tients with macular pseudo-operculum. Am J Oph- atrophic changes. Histopathologi- Department of Ophthalmology, thalmol. 1994;117:13-18. cally, the splitting of the peripheral Vanderbilt University, Medical Cen- 12. Guyer DR, Green WR. Idiopathic macular holes retina occurs superficially between ter East, Nashville, TN 37232-8808 and precursor lesions. In: Franklin RM, ed. Pro- the ganglion cell layer and the (e-mail: Donald.Gass@mcmail- ceedings of the Symposium on Retina and Vitre- nerve fiber layer. There is histo- ous, New Orleans Academy of Ophthalmology, .vanderbilt.edu). New Orleans, La. New York, NY: Kugler Publica- pathological and ultrastructural tions; 1993:135-162. evidence that degeneration of the REFERENCES 13. Yanoff N, Rahn EK, Zimmerman LE. Histopathol- inner portion of Mu¨ ller cells may ogy of juvenile retinoschisis. Arch Ophthalmol. be the cause of X-linked juvenile 1968;79:49-53. 13,14 1. Bron AJ, Tripathi RC, Tripathi BJ. Wolffe’s Anatomy 14. Condon GP, Brownstein S, Wang NS, et al. Con- retinoschisis. Unfortunately, of the Eye and . 8th ed. London, England: genital hereditary (juvenile X-linked) retinoschisis: none of the eyes examined patho- Chapman & Hall Medical Publishers; 1997:458. histopathologic and ultrastructural findings in three logically had typical foveomacular 2. Spencer WH. Ophthalmic Pathology: An Atlas and eyes. Arch Ophthalmol. 1986;104:576-583.

From the Archives of the ARCHIVES

A look at the past...

ARLOW’S case was an unmarried woman of thirty years who suf- fered from irregular menstruation. Has had sick headaches all her Mlife, much severer since she was fourteen years old. She had also other attacks, epileptic in character, becoming rigid with some convulsive move- ments and loss of consciousness for hours. After a correction of her refraction () () byR+1. +0.25 60°, L + 0.25 + 0.75 100°, the attacks ceased. This was five years ago, and there has been no recurrence.

Reference: Arch Ophthalmol. 1897:26:285-286.

ARCH OPHTHALMOL / VOL 117, JUNE 1999 823

©1999 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021