The Ophthalmoscope in the Lifetime of Hermann Von Helmholtz

SPECIAL ARTICLE The Ophthalmoscope in the Lifetime of Hermann von Helmholtz

C. Richard Keeler

n the whole history of medicine there is no more beautiful episode than the invention of the ophthalmoscope, and physiology has few greater triumphs.” Thus wrote American ophthalmologist Edward Loring1 in the opening paragraph of his Textbook of Ophthalmology in 1892, 2 years before the death of Hermann “Ivon Helmholtz.

On the 150th anniversary of the inven- from Heidelberg, where he had been a pro- tion or “discovery” of the ophthalmo- fessor for 3 years. scope (Figure 1) by Helmholtz, we have Helmholtz told the well-known story an opportunity, once again, to laud this of how 43 years before, he realized that he outstanding physicist of the 19th cen- had solved the mystery of how to pen- tury, just as his peers did on the 10th, 50th, etrate the black pupil of the eye and view and centennial years of this greatest of oph- the fundus. In 1851, Helmholtz had pub- thalmological inventions. lished his Beschreibung eines Augen- Prior to his invention, ophthalmolo- Spiegels,2 which gave a full account of the gists could not view the posterior section optical principles involved. Later he of the eye and struggled to explain cer- proudly told his father that he had re- tain classes of eye disease in which there ceived 18 orders for his instrument from was a dimness or loss of vision. Suddenly throughout Europe. in 1851, the world of ophthalmology was By the time of his death, a great num- taken by surprise, and a new epoch be- ber of different ophthalmoscopes had ap- gan. Immediately after the discovery of peared, many of them designed by the most the ophthalmoscope, men such as Albert eminent ophthalmologists of the day. Lor- von Graefe in Berlin, Germany, Edward ing, in the introduction to his textbook, Jaeger in Vienna, Austria, and William states, “[A]lmost every ophthalmologist has Bowman in London, England, started us- taken a hand in perfecting or at least alter- ing it. Every look into the eye became a ing the instrument, and from the first I have, discovery. perhaps, done more than my share.”1(p2) In 1893 at age 72 years, a year before There were no less than 140 instruments he died, Helmholtz went to America to at- on exhibition at a meeting hosted by Harry tend the World’s Fair in Chicago, Ill, as a Friedenwald and Casey Wood in Atlantic representative of the German govern- City, NJ, on the 50th anniversary of the in- ment. On his way back, he was invited to vention of the ophthalmoscope.3 New York, NY, by Professor Herman Helmholtz did not readdress the con- Knapp to inaugurate his new clinic be- struction of the ophthalmoscope but in- fore an audience of students and mem- stead became the greatest physicist and bers of the medical profession. Professor physiologist of the time. He had demon- Knapp, a former pupil who had begun strated that there were 3 essential ele- studying medicine in Germany the very ments to the working of an ophthalmo- year that the ophthalmoscope was in- scope: a source of illumination, a reflecting vented, had moved to New York in 1868 surface to direct light toward the eye, and a means of correcting an out-of-focus im- From the Royal College of Ophthalmologists Museum and Library, London, England. age on the fundus.

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 194

It was the lack of a strong, stable source of illumination that held up the development of the ophthalmoscope in the 19th century. Early users of the Helmholtz ophthalmoscope had to put up with a naked flickering candle as a light source. During this first decade, the candle was largely replaced by the oil lamp and then the paraffin- burning lamp. Various valiant attempts were made at this time to allow the source of illumination to follow the move- ments of the ophthalmoscope: the first by Ricardus Ulrich in 1854 with his candleholder precariously at- tached to the observation tube,4(p40) and the second by Lionel Beale in 1869 with his built-in oil lamp.5(p97) Figure 1. Early model of the Helmholtz ophthalmoscope, 1851. Late in the 18th century, Swiss- born physician and chemist Aime´ Ar- gand had invented a device that was to evolve into the most common source of illumination during the second half of the 19th century: the gas lamp. In the ordinary oil lamp, combustion was not complete. Ar- gand’s improvement was the replace- ment of the conventional wick with a ring. The flame became a hollow cylinder with a current of air ascend- ing through the inside so that the burning surface was doubled. Ar- gand’s brother accidentally discov- ered that a glass cylinder placed as a chimney over the flame steadied it, created a draft, and allowed the flame to yield the maximum amount of light.6 The gravity-fed oil lamp was fol- Figure 2. Nineteenth-century gravity-fed oil lamp (left) and gas lamp (right). lowed by the gas-burning lamp, which worked on the same prin- New York ophthalmologist, dem- ety of the United Kingdom, it never ciple (Figure 2); by 1869, this source onstrated this new technology at the went into production.8 Sir James of illumination had become the stan- American Ophthalmological Soci- McKenzie Davidson of Aberdeen, dard. This was fine for examinations ety when he presented the first oph- Scotland, who was one of the early in a fixed location such as an ophthalmoscope that used an electric pioneers of the use of x-ray in ophthalmologist’s office or the hospital, bulb.7 It was not a success, mainly thalmology, published an article in but for domiciliary visits the por- because of the unreliability and short the Lancet of January 1886 show- table candle or oil lamp, such as the life of the bulb. ing a diagram of an electric ophthal- one devised by Joseph Priestley Smith The following year saw the moscope.9 The third electric oph- of Birmingham, England,4(p270) was emergence of 3 ophthalmoscope de- thalmoscope was designed by Henry used well into the 20th century. These signs incorporating electric bulbs. Juler of London, England. Unlike lamps incorporated a reflecting mir- Like Dennett, Thomas Reid of Davidson’s concept, Juler’s instru- ror behind the candle and a strong Glasgow, Scotland, placed a bulb in- ment went into production. convex lens in front to condense the side the column of his instrument, Juler’s design (Figure 3) was light. but he used a prism instead of a mir- the attachment of a light source to In 1879, Thomas Edison was ror to project the light. Although this the outside of the ophthalmoscope working on the incandescent bulb. model was shown at the 1886 meet- body, close to the mirror with the Six years later William Dennett, a ing of the Ophthalmological Soci- bulb pointing toward the center ap-

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 195

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 the second model by Coccius,11(p42) but it did differ in 2 important re- spects: the mirror was convex and B was made of metal, not glass. Metal provided sharp edges to the sighthole with no restrictive viewing canal, which often occurred when using thick glass mirrors. Professor G. Theodor Ruete of Leipsig, Germany, was the first to in- troduce a concave mirror with aper- A ture, in 1853. This method of reflecting light toward the eye was to last the whole of Helmholtz’s lifetime and well beyond. Ruete used this concave focusing mirror in his new indirect ophthalmoscope (Figure 5), a method of ophthalmoscopy13 that C Helmholtz had anticipated and that was to prove of extraordinary impor- tance in ophthalmology. Most oph- Figure 3. Juler electric ophthalmoscope thalmoscopes thereafter accommo- showing the driving wheel (A) that moves the dated mirrors and lenses, allowing the lenses around the track, the lamp housing (B) containing the bulb, and the Le Clanche´ battery ophthalmologist to switch between (C) that powers the small lamp. direct and indirect ophthalmoscopy on the same instrument. erture. Apart from excessive sight- In 1854, Edward Jaeger of Vi- hole flare, this instrument had the Figure 4. The first ophthalmoscope by Adolf enna, Austria, introduced an oph- same problem of a short bulb life. Coccius, with a plano-mirror and biconvex lens thalmoscope with a choice of aper- for focusing the light. However, various models using this ture mirrors with a 4- or 7-in focal method became popular. length as well as his version of Helm- used only for direct ophthalmos- holtz’s glass plates, each of which METHODS OF copy, and the vogue for the indi- could be angled toward the source REFLECTING LIGHT rect method had caught on rapidly. of illumination.4(p102) Jaeger thereby His glass plates method of light combined the principles of Helm- Haynes Walton10 of St Mary’s Hos- reflection was quickly superceded holtz, Ruete, and others in one rather pital in London wrote in his Practi- when Epkens, an instrument maker complicated instrument. cal Treatise on the Diseases of the Eye in Holland, followed by others, used A common problem in all oph- in 1875, “The modes of reflecting a plano-mirror with an oval-shaped thalmoscopes when tilting the mir- light into the eye are manifold. It is aperture cut out of the middle. Ep- ror toward the light was having to done by unsilvered and by silvered kens then collaborated with Frans view the fundus obliquely through reflectors; by reflectors alone and by Donders of Utrecht to produce sev- the correcting lens, as the mirror was reflectors in combination with eral table-mounted model ophthal- mounted flat against the Rekoss disc. lenses; by prisms also and by lenses moscopes.11(p32-37) Four years previ- This produced a significant reduc- which are themselves silvered.” ously in 1847, Charles Babbage, of tion in vision and shift of the image Helmholtz brilliantly devised a calculating-machine fame, had used when viewing with higher-power method of light reflection by plac- the same mirror design in his pro- lenses. ing superimposed plates of plain totype ophthalmoscope.12 John Couper of the Royal Lon- glass at an angle to the incident light. The plano-mirror was an inef- don Ophthalmic Hospital, Moor- In his monograph,2 he was very pre- ficient gatherer of light. In an at- fields, overcame this problem in cise about the angle of incidence tempt to concentrate the source of 1875. He mounted his tiltable mir- needed to obtain the maximum il- illumination, Adolf Coccius in 1853 ror on a vertical axis so that the lumination when using either 1, 3, used a biconvex condensing lens on source of light could be received by or 4 plates, with partial polariza- an adjustable arm to focus the source the mirror, which was placed left or tion of the distractingly bright cor- of light onto the plano-mirror right at an angle to the Rekoss disc. neal reflex being achieved with 3 or (Figure 4). Although his first model The observer was able to view the more plates. His weak-light mirror, used a plano-mirror, later this be- fundus looking perpendicularly as it was also known, had its advo- came concave, further increasing the through whichever lens was appro- cates, especially those who wanted concentration of the light. priate. Couper’s first instrument, to detect slight variations and shades In 1854, Professor Karl von Ze- now in the Institute of Ophthalmolo- of color in the fundus. However, the hender constructed an instrument gy’s collection in London (Figure 6, Helmholtz ophthalmoscope could be that had an uncanny resemblance to left), may be his own unique de-

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 196

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Figure 7. Loring ophthalmoscope with a Figure 5. G. Theodor Ruete’s indirect ophthalmoscope using a concave mirror with aperture (3). rectangular concave tilting mirror.

sworth of Boston, Mass, who designed an angled concave mirror mounted in a disc that could be placed left or right on the head of a Loring ophthalmoscope.4(p346) This conveniently overcame the paral- lax in earlier instruments and moved the sight-hole and the mirror aperture closer together, allowing the observer to obtain a wider field of view of the fundus. In the same year, Edward Lor- ing went further than Couper and Wadsworth and produced his vertically held, tilting, rectangular concave mirror (Figure 7). With the sides of a round mirror cut off, a suf- ficient tilting angle could be achieved while keeping a minimal distance between the sight-hole and mirror aperture. This mirror was extremely simple to use in either the left or right position. Its construction became very popular and was to be incorporated into many other designs during the next 20 years. Until 1882, the need to use mirrors of more than one focal length for the direct and indirect method of ophthalmoscopy meant the re- Figure 6. John Couper’s first ophthalmoscope, with a slide-in Rekoss disc behind a tilting mirror (left). moval of one mirror for another. In Right, second model. that year, however, George Lind- say Johnson of London introduced sign. Couper’s next model (Figure ror was a turning point in the evo- an ophthalmoscope (Figure 8) with 6, right) had a more convenient and lution of the ophthalmoscope.14 2 mirrors fixed to an arm that could sophisticated method of swinging the Couper’s idea was followed in be rotated to bring the appropriate mirror left and right. His tilting mir- 1876 by one from Oliver Wad- mirror into position in front of the

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 197

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Figure 8. Lindsay Johnson ophthalmoscope with 2 mirrors for direct and indirect ophthalmoscopy.

sight-hole. The smaller mirror, with a 3-in focal length, rotated around its own axis, and the larger mirror had a focal length of 18 in. Later variations of 3 or even 4 mirrors, back to back, were to dominate the more sophisticated devices such as Figure 9. Miniature model of the Liebreich ophthalmoscope complete with clip for holding correcting the Morton ophthalmoscope for the lenses and 2 indirect condensing lenses. next 40 years. cave lenses. Liebreich was not the In 1873, shortly after arriving METHODS OF CORRECTION first to use this idea; 2 years before, in New York, Herman Knapp, a pro- Coccius had done the same.4(p33) lific inventor of surgical and diag- Shortly after Helmholtz produced his Jaeger’s 1854 ophthalmoscope nostic instruments, devised an al- first ophthalmoscope, a major defi- included 8 concave and 4 convex ternative and more efficient way of ciency became apparent. There was lenses, with the chosen lens having introducing a wider range of lenses no convenient means of correcting to be individually inserted into an in an ophthalmoscope.16 He was es- for presbyopia or refractive errors. aperture at the back of the instru- pecially interested in carrying out re- The optics of the Helmholtz instrument, a tedious and impractical pro- fractions and wanted not only a ment produced a converging image cedure. In 1869, a degree of sophis- wider range of lenses but smaller as the rays entered the observer’s tication crept in with Loring’s first of jumps in power. He used 2 Rekoss eye15; this was the main reason that several ophthalmoscope designs. This discs, one with convex lenses and the the early instruments used only con- instrument had 3 Rekoss discs, each other with concave lenses, so that the cave lenses. In 1852, Egbert Re- with 8 lenses: one disc contained con- bottom of one and the top of the koss, a university machinist who had cave lenses of moderate power, another overlapped. Knapp had ad- made Helmholtz’s original instru- other had convex lenses of low power, mired Loring’s 1869 model but ment, provided the essential break- and a third was composed of high di- wanted to avoid the loss of time and through by adding 2 rotatable discs, optric power, both convex and con- the tedious process of assembly and each containing a series of lenses. cave. This increased battery of lenses disassembly for each patient. Richard Liebreich, while in Ber- allowed Loring to perform refrac- Xavier Galezowski of Paris, lin, Germany, at von Graefe’s clinic tions as well as ophthalmoscopy. The France, who had invented his tubu- in 1855, designed a simple and very ophthalmoscope illustrated in lar indirect ophthalmoscope in 1862, popular ophthalmoscope (Figure 9) Figure 10 has a fourth disc with a designed a very different ophthal- with a clip mounted on an adjust- further combination of low convex moscope (Figure 11) 20 years later able arm behind the mirror for a se- and concave lenses that could have in 1882. This used a single Rekoss ries of unmounted convex and con- been tailor-made for the operator. disc but with 2 concentric rings of

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 198

lenses mounted within it. The outer vice was to become the standard ring had 19 concave lenses, and the chain-of-lenses design for the next inner one had 13 convex lenses. century. Pushing the disc up or pulling it In the attempt to automate the Figure 11. Galezowski ophthalmoscope down brought the chosen circle of introduction of a wide range of se- (dual-mirror side) with a Rekoss disc of 2 convex or concave lenses to the cen- quential convex and concave lenses concentric rings of lenses in the down position ter of the viewing aperture. Al- in a conventionally shaped head, for concave lenses. though Loring had shown the way perhaps the Roth and Callan oph- in 1874, in his Rekoss disc, half of thalmoscopes are the most inge- of Helmholtz would not be com- each of the concentric circles had nious. Although both ophthalmolo- plete without mentioning stand- convex lenses, and the other half had gists published details of their mounted ophthalmoscopes and concave lenses.4(p330) instruments in 1864,18,19 Peter Cal- the binocular indirect ophthalmo- Another method of introduc- lan’s visit to Berlin most likely al- scope. ing correcting lenses was a design by lowed him to observe August Roth’s Many of the best-known oph- Edward Jackson of Denver, Colo, in design and then imitate it. Only in thalmologists used stand-mounted 1887.4(p361) He used the length of the detail do they differ. ophthalmoscopes like the Follin and ophthalmoscope head and column The change in lens power in the Liebreich designs for teaching and to incorporate 2 vertically sliding Roth ophthalmoscope (Figure 13), studying the fundus to produce at- lens racks, one on top of the other, diopter by diopter up to 47 diop- lases. The first fundus photograph, each with 5 lenses of convex and ters, is achieved by turning a ser- published by Oswalt Gerloff of concave powers that could be used rated wheel that drives a peg along Go¨ttingen, Germany, in 1891,20 had singly or in combination. a wormlike groove, which rotates the been preceded by meticulously In the search for more lenses for Rekoss disc and linked quadrant. painted fundi showing a multi- refraction, John Couper designed the The peg is connected to a pointer on tude of pathological conditions that first “chain-of-lenses” ophthalmo- the outside that indicates the lens the ophthalmoscope could now scope (Figure 12).17 This was a bril- power required. reveal. liant engineering feat, with the oph- The size of correcting lenses in For today’s ophthalmologist, thalmoscope containing no less than ophthalmoscopes varied enor- the use of the direct ophthalmo- 72 lenses, each mounted in a free- mously, from 8 mm in the Landolt scope is less frequent. However, the moving brass cell. A cogwheel at the ophthalmoscope down to 3 mm in binocular indirect ophthalmo- center of the instrument was used to others. One early unidentified oph- scope, first introduced in 1861 by the drive the chain of lenses around a thalmoscope even had oval-shaped Frenchman Felix Giraud-Teulon,21 groove in the handle and head, with lenses, presumably attempting to is now used as a standard piece of each lens coming to rest precisely in pack the maximum number of lenses diagnostic equipment. Giraud- the center of the viewing aperture. within the Rekoss disc yet allowing Teulon’s binocular indirect ophthal- The Couper ophthalmoscope the observer plenty of vertical move- moscope, which used a solid rhom- was the forerunner of Andrew Stan- ment (Figure 14). boid prism to divide the image, had ford Morton’s first ophthalmo- A review of the development of a fixed interpupillary setting. His in- scope, in 1884.5(p119) The Morton de- the ophthalmoscope in the lifetime strument was followed in 1862 by

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 199

Figure 15. Photogravure portrait of Hermann von Helmholtz in 1894, the year of his death.

to the side of the patient. These instruments were difficult to use and quickly went out of fashion. The popularity of binocular indirect ophthalmoscopy had to wait many years until the introduction of Charles Schepens’ instrument in 1947, which incorporated bright and reliable illumination.22 The invention of the ophthalmoscope by Hermann von Helm- holtz was enormously exciting for the ophthalmologists of the day and fostered respect for and recogni- Figure 13. Roth ophthalmoscope with cover removed to show the internal groove and peg mechanism tion of ophthalmology as a medical for changing lens power. speciality. John Hughlings Jackson, a renowned English ophthalmologist and neurologist, never tired of impressing on physicians the value of the routine use of the ophthalmoscope. He stated that the physician was as much indebted to Helm- holtz as the ophthalmologist.23 Eight years after the discovery of the ophthalmoscope, Albrecht von Graefe, the greatest advocate of this new tool, presented Helmholtz with a cup at the 1858 Heidelberg Oph- thalmological Congress. The cup was inscribed with words that are as appropriate today as they were then: “To the creator of a new science, to the benefactor of mankind, in thank- ful remembrance of the invention of the ophthalmoscope”24 (Figure 15).

Accepted for publication September 19, Figure 14. Unidentified ophthalmoscope, circa 1870, with cover (right) removed to expose large 2001. oval-shaped lenses in the Rekoss disc. Corresponding author and re- prints: C. Richard Keeler, 1 Brook- a more sophisticated design by J. Both instruments were hand- field Park, London NW5 1ES, En- Zachariah Laurence and C. Heisch held and used for their light source gland (e-mail: rkeeler@freenetname of London.5(p95) an oil or gas lamp placed behind and .co.uk).

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 200

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 8. Rucker WC. A History of the Ophthalmoscope. 17. Couper J. A magazine refraction ophthalmoscope. REFERENCES Rochester, Minn: Whiting Printers; 1971. Trans Ophthalmol Soc U K. 1883;3:297-303. 9. Davidson JM. The electric light applied to the oph- 18. Callan PA. A new ophthalmoscope. Med Rec 1. Loring EG. Text-Book of Ophthalmology, Part 1. thalmoscope. Lancet. January 1886:209-210. (N Y). 1894;45:575. London, England: Henry Kimpton; 1892. 10. Walton H. A Practical Treatise on the Diseases of 19. Roth A. Ein Augenspiegel mit neuen Mechanismus 2. von Helmholtz HLF. Beschreibung eines Augen- the Eye. 3rd ed. London, England: Churchill; 1875. zur Selbsthattigen Linsenauswechslung. Klin Mon- Spiegels. Berlin, Germany: A Fo¨rstner’sche Ver- 11. den Tonkelaar I, Henkes HE, Van Leersum GK, eds. atsbla¨tter Augenheilkd. 1894;32.Jg.5:256-263. lagsbuchhandlung; 1851. Eye and instruments. In: Nineteenth Century Oph- 20. Gerloff O. Ueber die Photographie des Augenhin- 3. Friedenwald H. The history of the invention and thalmological Instruments in the Netherlands. Am- tergrundes. Klin Monatsbla¨tter Augenheilkd. 1891; of the development of the ophthalmoscope. JAMA. sterdam, the Netherlands: Batavian Lion; 1996. 29:397-403. 1902;38:549-552. 12. Wharton Jones T. Report on the ophthalmo- 21. von Haugwitz T. Optical instruments. Vol 2. In: 4. Schett A. The ophthalmoscope. Vol 1 [textbook]. scope. Br Foreign Med-Chir Rev. 1854;14:549. Hirschberg History of Ophthalmology. Bonn, In: Hirschberg History of Ophthalmology: The 13. Reute CGT. Der Augenspiegel und das Optom- Germany: J-P Wayenborgh Verlag; 1986:1-30. Monographs. Vol 2. Ostend, Belgium: J-P Way- eter. Go¨ttingen, Germany: Verlag der Dieterichs- 22. Schepens C. A new ophthalmoscope demonstra- enborgh; 1996:40. chen Buchhandlung; 1852. tion. Trans Am Acad Ophthalmol Otolaryngol. 5. Keeler CR. The ophthalmoscope. Vol 2 [atlas]. In: 14. CouperJ.Onadescriptionofanewophthalmoscope. 1947;51:298-301. Hirschberg History of Ophthalmology: The Mono- R Ophthalmic Hosp Rep. 1882;10:56-61. 23. Collins ET. The commencement of “The Ophthal- graphs. Vol 2. Ostend, Belgium: J-P Wayen- 15. Duke-Elder Sir WS. The examination of the eye. mic Hospital Reports.” In: The History and Tra- borgh; 1996:97. In: System of Ophthalmology. London, England: ditions of the Moorfields Eye Hospital. London, 6. Chambers Encyclopaedia. Vol 1. London, En- Henry Kimpton; 1962:233-363. England: HK Lewis & Co Ltd; 1929:119-131. gland: William & Robert Chambers; 1901. 16. Knapp H. Ophthalmoscopic optometry and de- 24. McKendrick JG. Hermann Ludwig Ferdinand von 7. Dennett WS. The electric light ophthalmoscope. scription of a new ophthalmoscope. Arch Ophal- Helmholtz: Masters of Medicine. London, En- Trans Am Ophthalmol Soc. 1885;4:156-157. mol Otolaryngol. 1874;3:1-25. gland: T Fisher Unwin; 1899.

IN OTHER JOURNALS

LANCET Microvascular Abnormalities and Incident Stroke: The Atherosclerosis Risk in Communities Study Tien Yin Wong, MD, MPH; Ronald Klein, MD, MPH; David J. Couper, PhD; Lawton S. Cooper, MD, MPH; Eyal Shahar, MD, MPH; Larry D. Hubbard, MAT; Marion R. Wofford, MD, MPH; and A. Richey Sharrett, MD, DrPH Background: Retinal microvascular abnormalities reflect damage from hypertension and other vascular processes. We ex- amined the relation of retinal microvascular abnormalities to incident stroke. Methods: A cohort of 10,358 men and women (aged 51 to 72 years) living in 4 communities had retinal photography and standardized grading for retinal microvascular abnormalities. In addition, the calibers of all retinal arterioles and venules were measured after digital conversion of the photographs, and a summary arteriole-to-venule ratio (AVR) was calculated as an index of arteriolar narrowing (with smaller AVR indicating greater narrowing). Incident hospitalized strokes were iden- tified from this cohort, and validated by case record reviews. Findings: Over an average of 3.5 years, 110 persons had incident strokes. After adjustment for age, gender, race, 6-year mean arterial blood pressure, diabetes, and other stroke risk factors, most retinal microvascular characteristics were predictive of incident stroke, with adjusted relative risks (and 95% confidence intervals) of 2.6 (1.6 - 4.2) for any retinopathy, 3.1 (1.7 - 5.7) for microaneurysms, 3.1 (1.4 - 6.7) for soft exudates, 2.6 (1.3 - 5.1) for blot hemorrhages, 2.3 (1.0 - 5.1) for flame- shaped hemorrhages, and 1.6 (1.0 - 2.5) for arterio-venous nicking. Relative risk of stroke increased with decreasing AVR (p=0.03). The associations were similar for ischemic strokes specifically, and for strokes in persons with hypertension, either with or without diabetes. Interpretation: Retinal microvascular abnormalities are related to incident stroke. The findings support a microvascular role in the pathogenesis of stroke, and suggest that retinal photography may be useful for cerebrovascular risk stratification in appropriate populations. (2001;358:1134-1140)

(REPRINTED) ARCH OPHTHALMOL / VOL 120, FEB 2002 WWW.ARCHOPHTHALMOL.COM 201