- Email: [email protected]
Hypertensive Retinopathy
Hypertensive Retinopathy Description, Classification, and Prognosis JOSEPH B. WALSH, MD
Abstract: In 1898 Marcus Gunn described the changes in retinal vessels noted with hypertension. Arteriolar narrowing, caliber irregularity, alterations of the light reflex, and hiding of the arterial blood column were noted. Arteriovenous crossing changes and capi lIary bed abnormalities, such as cotton-wool spots, retinal hemorrhages, and retinal edema were also mentioned, as well as blurred discs. In the 83 intervening years, little has been added to the description of hypertensive retinopathy, but our understanding has increased. Retinal vessels respond to elevations of systemic blood pressure by generalized arteriolar constriction. This can lead to arteriolar necrosis, retinal edema, cottonwool spots, hemorrhage, and disc edema. If the blood pressure is controlled, or slow rising, or if arteriolar sclerosis is present in the retinal arteries, then a picture of arteriolar irregularity will be noted and, depending upon the ability of the retinal vessels to contract, segmental constriction will be seen. In separating hypertensives from nonhypertensives, the most consistent ophthalmoscopic finding is arteriolar narrowing with focal irregularity. In prognosticating for survival, the best method available is the Keith-Wagener-Barker classification. However, the difficulty in separating Groups 1 and 2 of this classification has lead to numerous modifications that make comparisons from one study to another difficult. [Key words: arteriolar sclerosis, arteriovenous crossing changes, hypertensive retinopathy, Keith-Wagener-Barker, systemic hypertension, venous impedance.] Ophthalmology 89: 1127 -1131, 1982
In 1898, Marcus Gunn presented to the Ophthalmologic Society of the United Kingdom a series of observations he had made on the retinal vessels in patients with evidence of cerebral vascular insufficiency and/or renal disease.! The changes he enumerated were the narrowing and irregularity of retinal arteries. The irregularities could be due to focal constriction or From the Department of Ophthalmology, Montefiore Hospital and Medical Center/Albert Einstein College of Medicine, Bronx, New York. Presented at the Eighty-sixth Annual Meeting of American Academy of Ophthalmology, Atlanta, Georgia, November 1-6, 1981. Reprint requests to Joseph B. Walsh, MD, Montefiore Hospital and Medical Center/Albert Einstein College of Medicine, 111 East 210th Street, Bronx, NY 10467.
0161-6420/82/1000/1127/$1.05
© American Academy of Ophthalmology
dilation. They were also usually tortuous, had an increased light reflex (often irregular), and had a loss of their transparency (silver-wire). This last change could completely hide the blood column from view, but as Gunn observed, the lumen and blood flow could remain normal. At the site where these abnormal arteries crossed the vein, additional changes were noted. When the artery crossed the vein, the vein would become less apparent and even disappear on either side of the artery, appearing distally in a tapered fashion. The same changes, though less prominent, could be noted if the vein crossed the artery. The venous blood column could also be displaced in the direction of the arterial circulation. Impedance of venous flow distal to these aqnormal arteriovenous crossings was noted. Also described were patients with retinal hemorrhages, retinal white spots (cotton-wool spots), macu1127
OPHTHALMOLOGY. OCTOBER 1982 • VOLUME 89 • NUMBER 10
lar edema, macular star, and blurring of the optic disc margins. In the 83 years since Gunn's presentation, the ophthalmoscopic descriptive picture of hypertensive retinopathy has not been significantly altered. The emphasis has been on the understanding of such changes and how these changes correlate with systemic blood pressure and the resultant vascular changes in other target organs. The ocular fundus picture in hypertension is related closely to the status of the retinal arteries and the rate of rise and degree of systemic hypertension. The first major attempt to relate retinal vascular changes to survival in the hypertensive population was in 1939 by Keith, Wagener, and Barker. 2 They divided 219 hypertensive patients they had followed for at least five years into four groups on the basis of the ophthalmoscopic characteristics of each group (Table O. This grouping correlated directly with the degree of systemic hypertension and, inversely, with the prognosis for survival. In a later paper, Wagener, Clay, and Gipner presented more objective criteria for the classification of hypertensive retinopathy. 3 In grading arterial narrowing, they stressed that an objective guide should be used. This guide should be the examiner's memory of the size of normal retinal arterioles. Their grading for arterial narrowing is listed in Table 2. In a similar fashion, they graded focal arterial constriction and presented a method of grading arteriolar sclerosis. In addition, this report presented a classification of vascular hypertension and the retinal changes associated with each group (Table 3). By this method, the authors attempted to grade the acute and chronic retinal vascular changes due to hypertension and to do it in a precise and reproducible fashion. In spite of the effort by Wagener, Clay, and Gipner to bring precision to the classification of hypertensive retinopathy, application of this system to a clinical setting proved difficult to use and duplicate. In an attempt to bring a new and more complete study of hypertensive retinopathy, in 1953 Scheie graded changes of hypertension and arteriolar sclerosis separately.4 Hypertensive changes were defined as those related to arteriolar constriction. The vascular changes resulting from long-standing hypertension were defined by Scheie as arteriolar sclerosis (Table 4). Using Table 1. Ophthalmoscopic Grouping-Keith-Wagener-Barker*
Table 2. Wagener-Clay-Gipner Modification of Generalized Arteriolar Narrowing * Group 1 Reduction of caliber of arterioles to 3/4 of average caliber Group 2 Reduction of caliber of arterioles to 1/2 of average caliber Group 3 Reduction of caliber of arterioles to 1/3 of average caliber Group 4 Arterioles thread-like or invisible * Modified from reference 3.
these changes, he separately graded hypertensive changes and arteriolar sclerosis using a 0-4 scale (Table 5). This classification failed to account for the fact that arteriolar sclerosis may occur in the aging process and may be unrelated to hypertension. Leishman presented his method of classification in 1957, which stressed that the fundus picture in systemic hypertension depended upon the presence or absence of arteriolar sclerosis in the retinal vessels. 5 He showed that although hypertension may accelerate arteriolar sclerosis, arteriolar sclerosis may develop as part of the natural aging process. Arteriolar sclerosis was found with increasing frequency in the older population but varied from one individual to another. Many other factors (in addition to hypertension) may augment the aging changes of arteriolar sclerosis. These factors include ocular trauma, inflammation, and heredodegenerative disorders, and systemic diseases such as diabetes mellitus. Using this information, he established a seven-part classification into which he was able to place the group of normal and hypertensive patients he was studying (Table 6). With this seven-part grouping, Leishman emphasized the presence of involutional sclerosis as modifying the ability of the retinal arterioles to respond to hypertension. In patients with arteriolar sclerosis, the loss of the ability of the arteries to constrict resulted in generalized or focal dilation. The non-muscular arteriolar segments dilate secondarily to the increased Table 3. Wagener-Clay-Gipner Classification of Vascular Hypertension * Vascular Hypertension Neurogenic Acute (angiospastic)
Mild generalized arteriolar narrowing Generalized arteriolar narrowing and focal constriction; retinal and disc edema; cotton-wool spots; retinal hemorrhages
Chronic nonprogressive Chronic progressive
Grade 1 and 2 arterial narrowing Generalized and focal arterial narrowing; generalized arteriolar sclerosis
Terminal malignant
Papilledema; arterioles narrow with focal constriction
Group 1 Mild to moderate narrowing or sclerosis of the arteries Group 2 Moderate to marked sclerosis of the retinal arterioles; exaggeration of the light reflex; arteriovenous compression changes or generalized and/or localized narrowing of the arterioles Group 3 Retinal arteriolar narrowing and focal constriction; retinal edema; cotton-wool spots; hemorrhage Group 4 Group 3 plus papilledema * Modified from reference 2.
1128
Retinal Vascular Changes
* Modified from reference 4.
WALSH • HYPERTENSIVE RETINOPATHY
Table 4. * Ophthalmoscopic changes of hypertension 1. Narrowing of retinal arteries. No arteriolar sclerosis 2. Irregularity in arteriolar caliber due to localized spasm 3. Hemorrhages in retina due to capillary ischemia 4. Exudates in retina, including cotton-wool spots and edema residues 5. Disc and retinal edema 6. Arteriolar sclerosis Ophthalmoscopic changes of arteriolar sclerosis 1. Arteriolar light reflex changes 2. Arteriovenous crossing changes 3. Tortuosity of retinal vessels , Modified from reference 4.
intraluminal pressure, whereas the normal muscular segments respond normally by constricting. Since Leishman's monumental presentation, it has become accepted that the evaluation of hypertensive retinopathy must take into account the presence or absence of arteriolar sclerosis in interpreting the fundi of hypertensive patients. This classification, however, is somewhat complicated and has not gained wide popularity in daily clinical practice. . In experimental hypertension, many changes occur in the retinal capillary bed that may be examined by fluorescein angiography. 6,7 Using fluorescein angiography, the degree of perfusion in the capillary bed can be determined, as well as the integrity of the bloodretinal barrier. With a rapid rise in blood pressure and arteriolar constriction (autoregulation), the precapillary arterioles become occluded, and, with time, their smooth muscle cells necrose. With this loss of the muscle cell, the vessel loses its ability to remain constricted. Vessel dilatation is followed by plasma leakage into the vessel wall. Fluorescein angiography shows leakage due to breakdown of the blood-retinal barrier from vessel endothelial damage. With further leakage into the wall, the lumen becomes secondarily occluded. This leads to capillary occlusion with retinal edema, cotton-wool spots, and hemorrhages. Table 5. Scheie Classification Hypertension Grade 0 Grade 1 Grade 2 Grade 3 Grade 4
No changes Barely detectable arteriolar narrowing Obvious arteriolar narrowing with focal irregularities Grade 2 plus retinal hemorrhages and/or exudates Grade 3 plus papilledema
Arteriolar sclerosist Grade 0 Normal Grade 1 Barely detectable light reflex changes Grade 2 Obvious increased light reflex changes Grade 3 Copper-wire arterioles Grade 4 Silver-wire arterioles , Modified from reference 4.
t Grades 1-4 are accompanied by gradually increasing signs of arterio-
venous crossing changes.
Table 6. Leishman's Classification' Group Involutional sclerosis
Retinal Changes Arterioles: straight, narrow No arteriovenous crossing changes General picture of an aging fundus Arteriolar segment toward disc dilated Distal arteriolar segment narrow Major arterioles dilated (to normal size of youth) Arteriovenous crossing changes Normal Diffuse arteriolar constriction
2 Involutional sclerosis with hypertension 3 Advanced involutional sclerosis with hypertension 4 Normal fundus of youth 5 Early hypertension in youthful vessels 6 Fulminating hypertension Arterioles narrowed Papilledema and retinal edema Retinal hemorrhages Vein hidden at arteriovenous croSSing 7 Severe hypertension with Changes of Groups 2 and 5 relative sclerosis , Modified from reference 5.
The types of changes seen in hypertensive retinopathy have been reviewed, and the major historical classifications have been discussed. In evaluating an individual patient, however, it seems most reasonable to evaluate changes seen in a systematic manner. First examined are the arterial vessels, including their size, regularity, color, course, light reflex, and visiblity of the blood column. Unless the arterioles are narrowed dramatically, the most easily noted change in these vessels is irregularity of the caliber. Tortuosity of the arterioles may be seen on a congenital basis, where uniform tortuosity is present throughout the fundus. Segmental arterial tortuosity is always abnormal. The arteriovenous crossings should be evaluated for the degree of hiding of the venous blood column, as well as for changes in the course of the involved vein. These two changes, however, do not mean that there is impedance to the venous flow at this site. 8 To determine if impedance to flow is present at an arteriovenous crossing, examination of the retina distal to the crossing is required. Impedance is evident when the distal vein is darker, larger, and more tortuous than the proximal segment. Additional signs of impedance are capillary changes, such as dilation of the capillaries, and retinal hemorrhage, edema or cotton-wool spots. In frank obstruction, venous-venous collaterals may be noted. Visual fields and fluorescein angiography assist in evaluating venous impedance distal to arteriovenous crossings. The fields may show relative or absolute scotomas in areas of venous impedance, and the fluorescein angiogram may show turbulence at the crossing or highlight the abnormalities of the capillary bed listed above. The disc should be examined for edema. Also, the choroidal circulation may show changes related to hypertension, acute choroidal leakage, or even effusion, and chronic retinal pigment epithelial changes (Elschnig's spots or Siegrist's streaks) due to choroidal vascular insufficiency. 9
1129
OPHTHALMOLOGY. OCTOBER 1982 • VOLUME 89 • NUMBER 10
When each of the above parameters is evaluated, the examiner must compare them to the expected fundus picture in a similar age group of normal patients. Mild arteriolar light reflex changes and minimal arteriolar irregularity in a 70- year-old patient have a much lower chance of being associated with hypertension than do the same changes noted in a 20-year-old patient. In a young patient with recently diagnosed hypertension, irregular arteriolar calibar suggests long-standing or previous episodes of hypertension. Evidence of arteriolar necrosis (hemorrhage, cotton-wool spot, retinal edema, capillary nonperfusion and/or disc edema, and venous obstruction) is a sign that the retinal vascular system is responding to elevated blood pressure and not to the process of aging. The ophthalmologist, therefore, plays an important role in detecting hypertensives as well as in assisting in the monitoring of the control of the disease through serial evaluations of the fundus. Fundus photography is helpful in such assessments, as it allows temporal comparison. Whether the retinal vascular changes previously described can be used reliably to separate hypertensives from nonhypertensives is still debated. In a study using ophthalmoscopy and/or fundus photos, an attempt was made to divide patients into those with elevated blood pressure and those without hypertension by the retinal findings alone. 10 This study was designed so that there were comparisons, not only from one observer to another, but also from each observer unknowingly repeating evaluations on the same cases, thus serving as his own control. There was significant disagreement, not only between different observers' interpretations but also between those interpretations by the same observer. Even after careful evaluation and discussion to allow improvement of the identification of vascular changes, a group of seven ophthalmologists did not reliably identify AiV crossing changes but, at least, were more accurate in identifying sector arterial narrowing. In this study, retinal hemorrhages and exudates were too rare to be evaluated. This study points out the subjective aspect of ophthalmoscopy. Many studies use one observer to keep the findings consistent within that study. However, in comparing one study to another, objective criteria are not established easily. In a large prospective study of 855, 50-year-old men in Gothenburg, Sweden, followed for from four to ten years, it was found that attenuated retinal arterioles constituted the most sensitive ophthalmoscopic indicator of hypertension. 11 If focal arterial narrowing was also present, the probability of hypertension was increased. This study found that neither isolated broadened arteriolar light reflexes nor NV crossing changes discriminated between the hypertensive and the nonhypertensive population. In a smaller series of 100 patients, Stokoe felt that NV crossing changes enhanced his separation of the hypertensive patients from those who were not hypertensive. 12 In a study of 220 patients with coronary artery disease, retinal vessel changes were correlated with the 1130
presence or absence of hypertension. 13 The ocular findings that correlated with hypertension in ~he study were arterial narrowing and irregularity, and arteriovenous crossings. In addition to detecting hypertension, the ocular fundus has been used to predict the prognosis for survival within a hypertensive group. Keith-WagenerBarker's original study of untreated hypertensives showed a strikingly decreased rate of survival for each group compared to the preceding group. At three years after fundus examination, the survival by groups was: Group 1-70%; Group 2-62%; Group 3-22%; and Group 4-6%.2 Due to problems in separating groups 1 and 2, the Keith-Wagener-Barker grouping was modified when the original study was updated in 1966. 14 ,15 This modification graded arteriolar narrowing, both generalized and focal, and arteriolar sclerosis. Retinal hemorrhage, but not exudate, could be present in group 2. Table 7 presents the modified classification. Analysis of data on 540 patients, using this modified classification, showed that the duration of survival was lower in each group than in matched controls and was decreased when compared to the preceding group. Thus, the ten-year survival rate for each group was: Group 1-71%; Group 2-51%; Group 3-35%; and Group 4-21 %. In this study, the modified KeithWagener-Barker classification served as a helpful prognostic indicator in hypertensive cases. With better control of hypertension, survivorship has improved in all grades of retinopathy. 16 The ocular fundus is an important target organ for detecting and, as well, monitoring hypertension. The classifications that have so far been proposed do not improve on the description of the retinal response to hypertension given by Marcus Gunn. In detecting hypertension, arteriolar narrowing and focal constrictions seem to be the most sensitive indicators in the absence of hemorrhages and disc edema. The modified Keith-Wagener-Barker grouping of hypertensive retinal vascular changes remains the most commonly used ophthalmoscopic prognostic indicator for grouping hypertensive patients, and it has gained worldwide acTable 7. Modified Keith-Wagener-8arker Classification* Group 2
3
4
Sclerosis 0-4 <1 1 or> 0-4 General arteriolar 0-4 0-4 0-4 0-4 narrowing Focal arteriolar 0-4 0-4 0-4 0-4 narrowing Hemorrhage ± ± ± Exudate + ± Papilledema + 0-4 depends on amount or extent of vascular change + present - absent *
Modified from reference 15.
WALSH • HYPERTENSIVE RETINOPATHY
ceptance. The difficulty in defining reproducible criteria for Groups 1 and 2 has resulted in individual standards each time a comprehensive study is attempted.
9.
10.
REFERENCES 1. Gunn M. On ophthalmoscopic evidence of general arterial disease. Trans Ophthalmol Soc UK 1898; 18:356-81. 2. Keith NM, Wagener HP, Barker NW. Some different types of essential hypertension: their course and prognosis. Am J Med Sci 1939; 19-7:332-43. 3. Wagener HP, Clay GE, Gipner JF. Classification of retinal lesions in the presence of vascular hypertension. Trans Am Ophthalmol Soc 1947; 45:57-73. 4. Scheie HG. Evaluation of ophthalmoscopic changes of hypertension and arteriolar sclerosis. Arch Ophthalmol 1953; 49:117-38. 5. Leishman R. The eye in general vascular disease: hypertension and arteriosclerosis. BrJ Ophthalmol1957; 41:641-701. 6. Ashton N. The eye in malignant hypertension. Trans Am Acad Ophthalmol Otolaryngol 1972; 76: 17 -40. 7. Garner A, Ashton N, Tripathi R, et al. Pathogenesis of hypertensive retinopathy. An experimental study in the monkey. Br J Ophthalmol 1975; 59:3-44. 8. Seitz R. The Retinal Vessels; Comparative Ophthalmoscopic
11.
12.
13.
14.
15.
16.
and Histologic Studies on Healthy and Diseased Eyes, translated by FC Blodi. St Louis: CV Mosby, 1964; 67-74. de Venecia G, Wallow I, Houser D, Wahlstrom M. The eye in accelerated hypertension. 1. Elschnig's spots in nonhuman primates. Arch Ophthalmol 1980; 98:913-8. Kagan A, Aurell E, Dobree J, et al. A note on signs in the fundus oculi and arterial hypertension: conventional assessment and significance. Bull WHO 1966; 34:955-60. Svardsudd K, Wedel H, Aurell E, Tibblin G. Hypertensive eye ground changes. Prevalence, relation to blood pressure and prognostic importance. The study of men born in 1913. Acta Med Scand 1978; 204:159-67. Stokoe NL. Fundus changes in hypertension-a long-term clinical study. In: Cant JS, ed. The William Mackenzie Centenary Symposium on The Ocular Circulation in Health and Disease. London: Kimpton, 1969; 117-35. O'Sullivan P, Hickey N, Maurer B, et al. Retinal artery changes correlated with other hypertensive parameters in a coronary heart disease case-history study. Br Heart J 1968; 30:556-62. Breslin DJ, Gifford RW Jr, Fairbairn JF II. Essential hypertension: a twenty-year follow-up study. Circulation 1966; 33:87-97. Breslin DJ, Gifford RW Jr, Fairbairn JF II, Kearns TP. Prognostic importance of ophthalmoscopic findings in essential hypertension. JAMA 1966; 195:335-8. Breckenridge A, Dollery CT, Parry EH. Prognosis of treated hypertension. Changes in life expectancy and causes of death between 1952 and 1967. Q J Med 1970; 39:411-429.
1131
Abstract: In 1898 Marcus Gunn described the changes in retinal vessels noted with hypertension. Arteriolar narrowing, caliber irregularity, alterations of the light reflex, and hiding of the arterial blood column were noted. Arteriovenous crossing changes and capi lIary bed abnormalities, such as cotton-wool spots, retinal hemorrhages, and retinal edema were also mentioned, as well as blurred discs. In the 83 intervening years, little has been added to the description of hypertensive retinopathy, but our understanding has increased. Retinal vessels respond to elevations of systemic blood pressure by generalized arteriolar constriction. This can lead to arteriolar necrosis, retinal edema, cottonwool spots, hemorrhage, and disc edema. If the blood pressure is controlled, or slow rising, or if arteriolar sclerosis is present in the retinal arteries, then a picture of arteriolar irregularity will be noted and, depending upon the ability of the retinal vessels to contract, segmental constriction will be seen. In separating hypertensives from nonhypertensives, the most consistent ophthalmoscopic finding is arteriolar narrowing with focal irregularity. In prognosticating for survival, the best method available is the Keith-Wagener-Barker classification. However, the difficulty in separating Groups 1 and 2 of this classification has lead to numerous modifications that make comparisons from one study to another difficult. [Key words: arteriolar sclerosis, arteriovenous crossing changes, hypertensive retinopathy, Keith-Wagener-Barker, systemic hypertension, venous impedance.] Ophthalmology 89: 1127 -1131, 1982
In 1898, Marcus Gunn presented to the Ophthalmologic Society of the United Kingdom a series of observations he had made on the retinal vessels in patients with evidence of cerebral vascular insufficiency and/or renal disease.! The changes he enumerated were the narrowing and irregularity of retinal arteries. The irregularities could be due to focal constriction or From the Department of Ophthalmology, Montefiore Hospital and Medical Center/Albert Einstein College of Medicine, Bronx, New York. Presented at the Eighty-sixth Annual Meeting of American Academy of Ophthalmology, Atlanta, Georgia, November 1-6, 1981. Reprint requests to Joseph B. Walsh, MD, Montefiore Hospital and Medical Center/Albert Einstein College of Medicine, 111 East 210th Street, Bronx, NY 10467.
0161-6420/82/1000/1127/$1.05
© American Academy of Ophthalmology
dilation. They were also usually tortuous, had an increased light reflex (often irregular), and had a loss of their transparency (silver-wire). This last change could completely hide the blood column from view, but as Gunn observed, the lumen and blood flow could remain normal. At the site where these abnormal arteries crossed the vein, additional changes were noted. When the artery crossed the vein, the vein would become less apparent and even disappear on either side of the artery, appearing distally in a tapered fashion. The same changes, though less prominent, could be noted if the vein crossed the artery. The venous blood column could also be displaced in the direction of the arterial circulation. Impedance of venous flow distal to these aqnormal arteriovenous crossings was noted. Also described were patients with retinal hemorrhages, retinal white spots (cotton-wool spots), macu1127
OPHTHALMOLOGY. OCTOBER 1982 • VOLUME 89 • NUMBER 10
lar edema, macular star, and blurring of the optic disc margins. In the 83 years since Gunn's presentation, the ophthalmoscopic descriptive picture of hypertensive retinopathy has not been significantly altered. The emphasis has been on the understanding of such changes and how these changes correlate with systemic blood pressure and the resultant vascular changes in other target organs. The ocular fundus picture in hypertension is related closely to the status of the retinal arteries and the rate of rise and degree of systemic hypertension. The first major attempt to relate retinal vascular changes to survival in the hypertensive population was in 1939 by Keith, Wagener, and Barker. 2 They divided 219 hypertensive patients they had followed for at least five years into four groups on the basis of the ophthalmoscopic characteristics of each group (Table O. This grouping correlated directly with the degree of systemic hypertension and, inversely, with the prognosis for survival. In a later paper, Wagener, Clay, and Gipner presented more objective criteria for the classification of hypertensive retinopathy. 3 In grading arterial narrowing, they stressed that an objective guide should be used. This guide should be the examiner's memory of the size of normal retinal arterioles. Their grading for arterial narrowing is listed in Table 2. In a similar fashion, they graded focal arterial constriction and presented a method of grading arteriolar sclerosis. In addition, this report presented a classification of vascular hypertension and the retinal changes associated with each group (Table 3). By this method, the authors attempted to grade the acute and chronic retinal vascular changes due to hypertension and to do it in a precise and reproducible fashion. In spite of the effort by Wagener, Clay, and Gipner to bring precision to the classification of hypertensive retinopathy, application of this system to a clinical setting proved difficult to use and duplicate. In an attempt to bring a new and more complete study of hypertensive retinopathy, in 1953 Scheie graded changes of hypertension and arteriolar sclerosis separately.4 Hypertensive changes were defined as those related to arteriolar constriction. The vascular changes resulting from long-standing hypertension were defined by Scheie as arteriolar sclerosis (Table 4). Using Table 1. Ophthalmoscopic Grouping-Keith-Wagener-Barker*
Table 2. Wagener-Clay-Gipner Modification of Generalized Arteriolar Narrowing * Group 1 Reduction of caliber of arterioles to 3/4 of average caliber Group 2 Reduction of caliber of arterioles to 1/2 of average caliber Group 3 Reduction of caliber of arterioles to 1/3 of average caliber Group 4 Arterioles thread-like or invisible * Modified from reference 3.
these changes, he separately graded hypertensive changes and arteriolar sclerosis using a 0-4 scale (Table 5). This classification failed to account for the fact that arteriolar sclerosis may occur in the aging process and may be unrelated to hypertension. Leishman presented his method of classification in 1957, which stressed that the fundus picture in systemic hypertension depended upon the presence or absence of arteriolar sclerosis in the retinal vessels. 5 He showed that although hypertension may accelerate arteriolar sclerosis, arteriolar sclerosis may develop as part of the natural aging process. Arteriolar sclerosis was found with increasing frequency in the older population but varied from one individual to another. Many other factors (in addition to hypertension) may augment the aging changes of arteriolar sclerosis. These factors include ocular trauma, inflammation, and heredodegenerative disorders, and systemic diseases such as diabetes mellitus. Using this information, he established a seven-part classification into which he was able to place the group of normal and hypertensive patients he was studying (Table 6). With this seven-part grouping, Leishman emphasized the presence of involutional sclerosis as modifying the ability of the retinal arterioles to respond to hypertension. In patients with arteriolar sclerosis, the loss of the ability of the arteries to constrict resulted in generalized or focal dilation. The non-muscular arteriolar segments dilate secondarily to the increased Table 3. Wagener-Clay-Gipner Classification of Vascular Hypertension * Vascular Hypertension Neurogenic Acute (angiospastic)
Mild generalized arteriolar narrowing Generalized arteriolar narrowing and focal constriction; retinal and disc edema; cotton-wool spots; retinal hemorrhages
Chronic nonprogressive Chronic progressive
Grade 1 and 2 arterial narrowing Generalized and focal arterial narrowing; generalized arteriolar sclerosis
Terminal malignant
Papilledema; arterioles narrow with focal constriction
Group 1 Mild to moderate narrowing or sclerosis of the arteries Group 2 Moderate to marked sclerosis of the retinal arterioles; exaggeration of the light reflex; arteriovenous compression changes or generalized and/or localized narrowing of the arterioles Group 3 Retinal arteriolar narrowing and focal constriction; retinal edema; cotton-wool spots; hemorrhage Group 4 Group 3 plus papilledema * Modified from reference 2.
1128
Retinal Vascular Changes
* Modified from reference 4.
WALSH • HYPERTENSIVE RETINOPATHY
Table 4. * Ophthalmoscopic changes of hypertension 1. Narrowing of retinal arteries. No arteriolar sclerosis 2. Irregularity in arteriolar caliber due to localized spasm 3. Hemorrhages in retina due to capillary ischemia 4. Exudates in retina, including cotton-wool spots and edema residues 5. Disc and retinal edema 6. Arteriolar sclerosis Ophthalmoscopic changes of arteriolar sclerosis 1. Arteriolar light reflex changes 2. Arteriovenous crossing changes 3. Tortuosity of retinal vessels , Modified from reference 4.
intraluminal pressure, whereas the normal muscular segments respond normally by constricting. Since Leishman's monumental presentation, it has become accepted that the evaluation of hypertensive retinopathy must take into account the presence or absence of arteriolar sclerosis in interpreting the fundi of hypertensive patients. This classification, however, is somewhat complicated and has not gained wide popularity in daily clinical practice. . In experimental hypertension, many changes occur in the retinal capillary bed that may be examined by fluorescein angiography. 6,7 Using fluorescein angiography, the degree of perfusion in the capillary bed can be determined, as well as the integrity of the bloodretinal barrier. With a rapid rise in blood pressure and arteriolar constriction (autoregulation), the precapillary arterioles become occluded, and, with time, their smooth muscle cells necrose. With this loss of the muscle cell, the vessel loses its ability to remain constricted. Vessel dilatation is followed by plasma leakage into the vessel wall. Fluorescein angiography shows leakage due to breakdown of the blood-retinal barrier from vessel endothelial damage. With further leakage into the wall, the lumen becomes secondarily occluded. This leads to capillary occlusion with retinal edema, cotton-wool spots, and hemorrhages. Table 5. Scheie Classification Hypertension Grade 0 Grade 1 Grade 2 Grade 3 Grade 4
No changes Barely detectable arteriolar narrowing Obvious arteriolar narrowing with focal irregularities Grade 2 plus retinal hemorrhages and/or exudates Grade 3 plus papilledema
Arteriolar sclerosist Grade 0 Normal Grade 1 Barely detectable light reflex changes Grade 2 Obvious increased light reflex changes Grade 3 Copper-wire arterioles Grade 4 Silver-wire arterioles , Modified from reference 4.
t Grades 1-4 are accompanied by gradually increasing signs of arterio-
venous crossing changes.
Table 6. Leishman's Classification' Group Involutional sclerosis
Retinal Changes Arterioles: straight, narrow No arteriovenous crossing changes General picture of an aging fundus Arteriolar segment toward disc dilated Distal arteriolar segment narrow Major arterioles dilated (to normal size of youth) Arteriovenous crossing changes Normal Diffuse arteriolar constriction
2 Involutional sclerosis with hypertension 3 Advanced involutional sclerosis with hypertension 4 Normal fundus of youth 5 Early hypertension in youthful vessels 6 Fulminating hypertension Arterioles narrowed Papilledema and retinal edema Retinal hemorrhages Vein hidden at arteriovenous croSSing 7 Severe hypertension with Changes of Groups 2 and 5 relative sclerosis , Modified from reference 5.
The types of changes seen in hypertensive retinopathy have been reviewed, and the major historical classifications have been discussed. In evaluating an individual patient, however, it seems most reasonable to evaluate changes seen in a systematic manner. First examined are the arterial vessels, including their size, regularity, color, course, light reflex, and visiblity of the blood column. Unless the arterioles are narrowed dramatically, the most easily noted change in these vessels is irregularity of the caliber. Tortuosity of the arterioles may be seen on a congenital basis, where uniform tortuosity is present throughout the fundus. Segmental arterial tortuosity is always abnormal. The arteriovenous crossings should be evaluated for the degree of hiding of the venous blood column, as well as for changes in the course of the involved vein. These two changes, however, do not mean that there is impedance to the venous flow at this site. 8 To determine if impedance to flow is present at an arteriovenous crossing, examination of the retina distal to the crossing is required. Impedance is evident when the distal vein is darker, larger, and more tortuous than the proximal segment. Additional signs of impedance are capillary changes, such as dilation of the capillaries, and retinal hemorrhage, edema or cotton-wool spots. In frank obstruction, venous-venous collaterals may be noted. Visual fields and fluorescein angiography assist in evaluating venous impedance distal to arteriovenous crossings. The fields may show relative or absolute scotomas in areas of venous impedance, and the fluorescein angiogram may show turbulence at the crossing or highlight the abnormalities of the capillary bed listed above. The disc should be examined for edema. Also, the choroidal circulation may show changes related to hypertension, acute choroidal leakage, or even effusion, and chronic retinal pigment epithelial changes (Elschnig's spots or Siegrist's streaks) due to choroidal vascular insufficiency. 9
1129
OPHTHALMOLOGY. OCTOBER 1982 • VOLUME 89 • NUMBER 10
When each of the above parameters is evaluated, the examiner must compare them to the expected fundus picture in a similar age group of normal patients. Mild arteriolar light reflex changes and minimal arteriolar irregularity in a 70- year-old patient have a much lower chance of being associated with hypertension than do the same changes noted in a 20-year-old patient. In a young patient with recently diagnosed hypertension, irregular arteriolar calibar suggests long-standing or previous episodes of hypertension. Evidence of arteriolar necrosis (hemorrhage, cotton-wool spot, retinal edema, capillary nonperfusion and/or disc edema, and venous obstruction) is a sign that the retinal vascular system is responding to elevated blood pressure and not to the process of aging. The ophthalmologist, therefore, plays an important role in detecting hypertensives as well as in assisting in the monitoring of the control of the disease through serial evaluations of the fundus. Fundus photography is helpful in such assessments, as it allows temporal comparison. Whether the retinal vascular changes previously described can be used reliably to separate hypertensives from nonhypertensives is still debated. In a study using ophthalmoscopy and/or fundus photos, an attempt was made to divide patients into those with elevated blood pressure and those without hypertension by the retinal findings alone. 10 This study was designed so that there were comparisons, not only from one observer to another, but also from each observer unknowingly repeating evaluations on the same cases, thus serving as his own control. There was significant disagreement, not only between different observers' interpretations but also between those interpretations by the same observer. Even after careful evaluation and discussion to allow improvement of the identification of vascular changes, a group of seven ophthalmologists did not reliably identify AiV crossing changes but, at least, were more accurate in identifying sector arterial narrowing. In this study, retinal hemorrhages and exudates were too rare to be evaluated. This study points out the subjective aspect of ophthalmoscopy. Many studies use one observer to keep the findings consistent within that study. However, in comparing one study to another, objective criteria are not established easily. In a large prospective study of 855, 50-year-old men in Gothenburg, Sweden, followed for from four to ten years, it was found that attenuated retinal arterioles constituted the most sensitive ophthalmoscopic indicator of hypertension. 11 If focal arterial narrowing was also present, the probability of hypertension was increased. This study found that neither isolated broadened arteriolar light reflexes nor NV crossing changes discriminated between the hypertensive and the nonhypertensive population. In a smaller series of 100 patients, Stokoe felt that NV crossing changes enhanced his separation of the hypertensive patients from those who were not hypertensive. 12 In a study of 220 patients with coronary artery disease, retinal vessel changes were correlated with the 1130
presence or absence of hypertension. 13 The ocular findings that correlated with hypertension in ~he study were arterial narrowing and irregularity, and arteriovenous crossings. In addition to detecting hypertension, the ocular fundus has been used to predict the prognosis for survival within a hypertensive group. Keith-WagenerBarker's original study of untreated hypertensives showed a strikingly decreased rate of survival for each group compared to the preceding group. At three years after fundus examination, the survival by groups was: Group 1-70%; Group 2-62%; Group 3-22%; and Group 4-6%.2 Due to problems in separating groups 1 and 2, the Keith-Wagener-Barker grouping was modified when the original study was updated in 1966. 14 ,15 This modification graded arteriolar narrowing, both generalized and focal, and arteriolar sclerosis. Retinal hemorrhage, but not exudate, could be present in group 2. Table 7 presents the modified classification. Analysis of data on 540 patients, using this modified classification, showed that the duration of survival was lower in each group than in matched controls and was decreased when compared to the preceding group. Thus, the ten-year survival rate for each group was: Group 1-71%; Group 2-51%; Group 3-35%; and Group 4-21 %. In this study, the modified KeithWagener-Barker classification served as a helpful prognostic indicator in hypertensive cases. With better control of hypertension, survivorship has improved in all grades of retinopathy. 16 The ocular fundus is an important target organ for detecting and, as well, monitoring hypertension. The classifications that have so far been proposed do not improve on the description of the retinal response to hypertension given by Marcus Gunn. In detecting hypertension, arteriolar narrowing and focal constrictions seem to be the most sensitive indicators in the absence of hemorrhages and disc edema. The modified Keith-Wagener-Barker grouping of hypertensive retinal vascular changes remains the most commonly used ophthalmoscopic prognostic indicator for grouping hypertensive patients, and it has gained worldwide acTable 7. Modified Keith-Wagener-8arker Classification* Group 2
3
4
Sclerosis 0-4 <1 1 or> 0-4 General arteriolar 0-4 0-4 0-4 0-4 narrowing Focal arteriolar 0-4 0-4 0-4 0-4 narrowing Hemorrhage ± ± ± Exudate + ± Papilledema + 0-4 depends on amount or extent of vascular change + present - absent *
Modified from reference 15.
WALSH • HYPERTENSIVE RETINOPATHY
ceptance. The difficulty in defining reproducible criteria for Groups 1 and 2 has resulted in individual standards each time a comprehensive study is attempted.
9.
10.
REFERENCES 1. Gunn M. On ophthalmoscopic evidence of general arterial disease. Trans Ophthalmol Soc UK 1898; 18:356-81. 2. Keith NM, Wagener HP, Barker NW. Some different types of essential hypertension: their course and prognosis. Am J Med Sci 1939; 19-7:332-43. 3. Wagener HP, Clay GE, Gipner JF. Classification of retinal lesions in the presence of vascular hypertension. Trans Am Ophthalmol Soc 1947; 45:57-73. 4. Scheie HG. Evaluation of ophthalmoscopic changes of hypertension and arteriolar sclerosis. Arch Ophthalmol 1953; 49:117-38. 5. Leishman R. The eye in general vascular disease: hypertension and arteriosclerosis. BrJ Ophthalmol1957; 41:641-701. 6. Ashton N. The eye in malignant hypertension. Trans Am Acad Ophthalmol Otolaryngol 1972; 76: 17 -40. 7. Garner A, Ashton N, Tripathi R, et al. Pathogenesis of hypertensive retinopathy. An experimental study in the monkey. Br J Ophthalmol 1975; 59:3-44. 8. Seitz R. The Retinal Vessels; Comparative Ophthalmoscopic
11.
12.
13.
14.
15.
16.
and Histologic Studies on Healthy and Diseased Eyes, translated by FC Blodi. St Louis: CV Mosby, 1964; 67-74. de Venecia G, Wallow I, Houser D, Wahlstrom M. The eye in accelerated hypertension. 1. Elschnig's spots in nonhuman primates. Arch Ophthalmol 1980; 98:913-8. Kagan A, Aurell E, Dobree J, et al. A note on signs in the fundus oculi and arterial hypertension: conventional assessment and significance. Bull WHO 1966; 34:955-60. Svardsudd K, Wedel H, Aurell E, Tibblin G. Hypertensive eye ground changes. Prevalence, relation to blood pressure and prognostic importance. The study of men born in 1913. Acta Med Scand 1978; 204:159-67. Stokoe NL. Fundus changes in hypertension-a long-term clinical study. In: Cant JS, ed. The William Mackenzie Centenary Symposium on The Ocular Circulation in Health and Disease. London: Kimpton, 1969; 117-35. O'Sullivan P, Hickey N, Maurer B, et al. Retinal artery changes correlated with other hypertensive parameters in a coronary heart disease case-history study. Br Heart J 1968; 30:556-62. Breslin DJ, Gifford RW Jr, Fairbairn JF II. Essential hypertension: a twenty-year follow-up study. Circulation 1966; 33:87-97. Breslin DJ, Gifford RW Jr, Fairbairn JF II, Kearns TP. Prognostic importance of ophthalmoscopic findings in essential hypertension. JAMA 1966; 195:335-8. Breckenridge A, Dollery CT, Parry EH. Prognosis of treated hypertension. Changes in life expectancy and causes of death between 1952 and 1967. Q J Med 1970; 39:411-429.
1131