BRITISH MEDICAL JOURNAL
1443
2 JUNE 1979
Wadia, N H, in Tropical Neurology, ed J D Spillane, p 63. London, Oxford University Press, 1973. 4 Tandon, P N, and Pathak, S N, in Tropical Neurology, ed J D Spillane, p 37. London, Oxford University Press, 1973. 5 Medical Research Council Working Party on Tuberculosis of the Spine, Journal of Bone and Joint Surgery, 1976, 58B, 399. 6 Fox, W, Bulletin of the International Union against Tuberculosis, 1978, 53, 268. 7 Canetti, G, Debeyre, J, and Seze, S de, Revue de la Tuberculose, 1957, 21, 1337. 8 Dobson, J, Journal of Bone and Joint Surgery, 1951, 33-B, 517. 9Medical Research Council Working Party on Tuberculosis of the Spine, British Journal of Surgery, 1974, 61, 853. 10 Griffiths, D LI, Seddon, H J, and Roaf, R, Pott's Paraplegia. London, Oxford University Press, 1956. 15 Alexander, G L, Proceedings of the Royal Society of Medicine, 1945-1946, 39, 730. 12 Hodgson, A R, and Stock, F E, British Journal of Surgery, 1956, 44, 266. 3 Medical Research Council Working Party on Tuberculosis of the Spine, Tubercle, 1978, 59, 79. 3
Ocular complications of temporal arteritis
mean ESR increases by 3 mm/hour with every decade over the age of 30 years," but in temporal arteritis it is usually over 70 mm/hour. Since the arteritis is patchy, both circumferentially and longitudinally, a biopsy specimen from the temporal artery does not always show the typical multinucleated giant cells; but other changes, such as reduplication or fragmentation of the internal elastic lamina and round cell infiltration with intimal thickening, are more often seen. Treatment with steroids should be given at once-without waiting for the biopsy results-since recovery of vision is possible.'2 The initial treatment should be prednisone 15 mg four times daily, reducing as the ESR comes down to near normal limits, usually within three weeks. If the headache and other symptoms are not ameliorated within 48 hours the diagnosis should be reviewed. Treatment should be continued for a minimum of six months, the dosage being titrated against the clinical well-being of the patient and the ESR. Temporal arteritis is a self-limiting disease that lasts on average about a year,' but relapses have occurred with stopping treatment, even after several years. Cullen, J F, and Coleiro, J A, Survey of Ophthalmology, 1976, 20, 247. Huston, K A, et al, Annals of Internal Medicine, 1978, 88, 162. McLeod, D, British Journal of Ophthalmology, 1976, 60, 551. McLeod, D, and Kohner, E M, New England Journal of Medicine, 1977, 297, 1180. 5 McLeod, D, et al, British3ournal of Ophthalmology, 1978, 62, 591. 6 Winter, B J, Cryer, T H, and Hameroff, S B, Southern MedicallJournal, 1977, 70, 1479. 7Barricks, M E, et al, Brain, 1977, 100, 209. 8 Long, R G, Friedmann, A I, and James, D G, PoJstgraduate Medical 7ournal, 1976, 52, 689. 9Kansu, T, et al, Archives of Neurology, 1977, 34, 624. lo Gill, C R, Journal of the Kentucky Medical Association, 1977, 75, 483. 1 Hayes, G S, and Stinson, I N, Archives of Ophthalmology, 1976, 94, 939. 12 Parsons-Smith, B G, in Medical Ophthalmology, ed F Clifford Rose, chapter 23. London, Chapman and Hall, 1976. I 2
Early diagnosis is essential in temporal (giant-cell) arteritis to prevent blindness. The possibility must be considered in any patient over the age of 55 who presents with severe persistent headache, particularly if there are also systemic features such as malaise, anorexia, loss of weight, fever, night sweats-and especially pains in the limbs or joints that are worse in the morning and aggravated by exercise. Another valuable diagnostic pointer is pain in the masseter muscles on chewing. Examination will show temporal arteries that are tortuous, thickened, pulseless, and-most importanttender, so that the patient may say that combing her hair or wearing a hat has become impossible. Visual loss is the first complaint in nearly half of all cases': usually sudden, unilateral, permanent, and complete but less often gradual, bilateral, transient, and partial.2 The ophthalmoscopic findings are variable and depend on the vessels affected, but the most common pattern is ischaemic optic neuropathy due to posterior ciliary arteritis. In addition to narrowed vessels and a few circumpapillary haemorrhages, there may be oedema of the disc due to blocking of axoplasmic flow at the nerve head.3 This interference with axoplasmal transport is responsible for the transient cotton wool spots seen later on. A similar appearance develops in occlusion of the central retinal artery4 5; with complete occlusion the cotton wool spots accumulate at the optic nerve head, but with partial occlusion they occur at the junction of ischaemic and non-ischaemic retina, and are hence circumpapillary; they are not seen in the periphery. Owing to the pallor of the surrounding ischaemic retina a cherry red spot is seen in central retinal artery occlusion. The ocular vessels most often affected are the ophthalmic, central retinal, and short posterior ciliary arteries,6 but the long posterior ciliary arteries may also be damaged, resulting in the rare anterior segment ischaemia. Other uncommon ocular presentations include diplopia, which is often transient,7 and scleritis.8 The disease affects small and medium-sized vessels throughout the body and not just those of the head; the vessels most frequently found to be affected at necropsy are those containing elastic tissue, such as the extracranial vertebral arteries. The diagnosis is confirmed by finding a raised erythrocyte sedimentation rate (ESR), but this is not invariable.9 10 The
3 4
Coronary arteriography before aortic valve replacement The prognosis of patients with symptomatic aortic valve disease is poor. Without specific treatment the average life expectancy in those with aortic stenosis' is no more than three to four years. But replacement of the valve restores health and provides a good prospect of prolonged survival. The risks and results of surgery are largely determined by the state of the myocardium and its arterial supply, which cannot be decided on clinical grounds. Angina pectoris, for example, may be due to the valve disease alone, or to additional coronary artery disease.2 Conversely, coronary artery disease may be found in those without angina,3 4 albeit less commonly. Thus coronary artery disease can be identified only by coronary arteriography-a procedure that has a low risk of mortality or morbidity.5 6 Arteriography is clearly indicated if there is doubt about the severity of the aortic valve disease, but should it be routine ? In those centres where it has become a standard procedure the incidence of coronary artery disease in patients being assessed for aortic valve surgery has generally been found3 4 7to be between 20% and 3Oo%, but it may be8 as high as 48%. Coronary disease may need to be identified before replacement of the aortic valve for two reasons. Firstly, the operation
1444
may be made safer. In the 1593 operations for isolated replacement of the aortic valve in Britain in 1977 there were 135 hospital deaths-a mortality rate of 8-3%.9 This reflects surgical practice throughout the country, but there was considerable variation among centres: the rate in a good centre accepting all comers should be around 500.10 Most deaths are due to myocardial damage occurring during surgery, usually frank infarction.1' The standard method of protecting the myocardium during the operation has been to perfuse the coronary arteries with blood, and if the surgeon knows accurately the anatomy of the left main coronary artery he may be prevented from selectively cannulating one branch, thus causing underperfusion of the other.12 Nevertheless, coronary cannulation carries some risk, and local dissection of the arterial wall accounted for two of the deaths in the series quoted above." This and other reasons have persuaded cardiac surgeons to turn from coronary perfusion for minimising myocardial damage, and to use instead cardiac cooling with potassium-induced cardiac arrest. This method of myocardial preservation causes less damage'3 and is associated with a lower incidence of operative myocardial infarction'4-and there is no need for preoperative assessment of the coronary arteries. So the first justification for preoperative coronary arteriography is becoming obsolete. The second reason, however, is more cogent. Preoperative arteriography should help to improve the long-term results by identifying lesions in the coronary arteries that warrant bypass surgery. Replacement of the valve alone brings symptomatic relief in most patients with aortic valve disease,'5 as might be expected from a knowledge of the pathophysiology; but a few patients are left with persisting angina. Furthermore, patients with coexisting coronary artery disease do less well after surgery compared with those with normal coronary arteries.'6 So in suitable patients the combined operation would be desirable on both counts-were it not for the increased operative risk. This is difficult to measure. Some authors report no increased risk,'7 whereas others have found a threefold increase in the mortality rate.'8 In British units in 1977 the mortality rate for combined aortic valve replacement and coronary artery bypass grafting9 was 11.7%-higher than that for isolated aortic valve replacement, and likely to remain so because of the longer time taken for the operation. So we can draw no firm conclusions about the place of coronary arteriography before aortic valve replacement. Among the many unanswered questions are the functional importance of arteriographic narrowings in patients with aortic valve disease, the relative risks of isolated aortic valve surgery and the combined operation, and the long-term fate of vein grafts. For the time being cardiac centres have evolved their own policies: those using coronary cannulation rely on detailed knowledge of the anatomy of the arteries, while those using cardioplegic techniques undertake coronary arteriography in patients with angina and then in each individual decide the merits of bypass grafting. What does emerge from the confusion is the variation in hospital mortality rates among different centres in Britain, which (for elective aortic valve replacement) may be as low as 1%.19 Any hospital undertaking open heart surgery has a duty to make its results available to referring physicians so that they can take decisions about the surgical treatment of aortic valve disease on the basis of reliable
information. 1 2
Mitchell, A M, et al, American Heart_Journal, 1954, 48, 684. British MedicalJournal, 1977, 1, 597.
3 Swanton, R H, et al, British Heart_Journal, 1977, 39, 1347. 'Harris, C N, et al, British HeartJournal, 1975, 37, 656.
BRITISH MEDICAL JOURNAL
2 JUNE 1979
Pridie, R B, et al, British HeartJournal, 1976, 38, 1200. Lacy, J, et al, Annals of Thoracic Surgery, 1977, 23, 429. 7 Baxter, R H, et al, British Heart3Journal, 1978, 40, 918. 8 Linhart, J W, Journal of Thoracic and Cardiovascular Surgery, 1968, 55, 811. 9 United Kingdom Cardiac Surgical Register. London, Society of Thoracic and Cardiovascular Surgeons of Great Britain and Ireland, 1978. 10 Ross, J K, et al, British Medical_Journal, 1976, 2, 1485. 1 Sharratt, G P, Rees, P, and Conway, N, Journal of Thoracic and Cardiovascular Surgery, 1976, 71, 869. 12 Furlong, M B, et al,Journal of Thoracic and Cardiovascular Surgery, 1972, 63, 185. 13 Richardson, J V, et al, Circulation, 1979, 59, 75. 14 Ellis, R J, et al, Circulation, 1978, 58, suppl 1, 57. 15 Baker, C, and Somerville, J, British MedicalJournal, 1964, 1, 197. 16 Copeland, J G, et al, Journal of Thoracic and Cardiovascular Surgery, 1977, 74, 875. 17 McManus, Q, et al,3Journal of Thoracic and Cardiovascular Surgery, 1978, 75, 865. 18 Loop, F D, et al, Circulation, 1977, 55, 169. 19 Parker, D J, and Seabra-Gomes, R S, British HeartJ'ournal, 1974, 36, 1039. 5
6
Senile disciform macular degeneration The latest statistics' from the Department of Health show that macular lesions account for over 20% of the blind in the age group 65-74 and for over 45%0 in the over 75s. Presumably almost all of these persons have senile macular degeneration and some of them, in turn, have senile disciform macular degeneration. Through the ophthalmoscope, senile macular degeneration is seen as an area of mottling with small clumps of pigment, sometimes with yellow spots called drusen, centred roughly 21 disc diameters temporal to the optic disc. The area of degeneration extends over one or two disc diameters. The cause of senile macular degeneration is unknown, but its tendency sooner or later to become bilateral makes it a ,disabling condition: the central area of the patient's field of vision is progressively affected so that reading or recognition of faces gradually becomes impossible. There is no satisfactory treatment. Nevertheless, the sufferers are always encouraged to be told that the disease will not spread outwards beyond the central 10 degrees or so to blot out their peripheral field, which is so valuable for navigation, as in walking and crossing roads. Some patients with macular degeneration become blind suddenly as a result of the development of disciform lesions. Studies using intravenous fluorescein and retinal photography have shown that an important stage in the evolution of these lesions is the appearance of new small vessels beneath. the pigment epithelium of the macular area of the retina.2-9 The new vessels are fragile and may bleed, suddenly destroying the centre of the visual field: in such cases there is a large sheet of subretinal haemorrhage temporal to the optic disc. The ophthalmoscopic picture may be mistaken for a central or tributary retinal vein occlusion. The haemorrhage may spread into the retina and also sometimes into the vitreous, with the final lesion taking the form of an irregular lumpy plaque of fibrous tissue. At that stage little if any recovery can be expected and no treatment of that eye is possible. We have, however, accumulating evidence that argon laser photocoagulation can destroy the new subretinal vessels in their early stages and thereby prevent the disastrous haemorrhage, provided that the new vessels have not yet involved the foveola (the central portion of the fovea), where light coagulation would of itself destroy central vision.10-15 The new vessels may start at any point within the wider macular
1443
2 JUNE 1979
Wadia, N H, in Tropical Neurology, ed J D Spillane, p 63. London, Oxford University Press, 1973. 4 Tandon, P N, and Pathak, S N, in Tropical Neurology, ed J D Spillane, p 37. London, Oxford University Press, 1973. 5 Medical Research Council Working Party on Tuberculosis of the Spine, Journal of Bone and Joint Surgery, 1976, 58B, 399. 6 Fox, W, Bulletin of the International Union against Tuberculosis, 1978, 53, 268. 7 Canetti, G, Debeyre, J, and Seze, S de, Revue de la Tuberculose, 1957, 21, 1337. 8 Dobson, J, Journal of Bone and Joint Surgery, 1951, 33-B, 517. 9Medical Research Council Working Party on Tuberculosis of the Spine, British Journal of Surgery, 1974, 61, 853. 10 Griffiths, D LI, Seddon, H J, and Roaf, R, Pott's Paraplegia. London, Oxford University Press, 1956. 15 Alexander, G L, Proceedings of the Royal Society of Medicine, 1945-1946, 39, 730. 12 Hodgson, A R, and Stock, F E, British Journal of Surgery, 1956, 44, 266. 3 Medical Research Council Working Party on Tuberculosis of the Spine, Tubercle, 1978, 59, 79. 3
Ocular complications of temporal arteritis
mean ESR increases by 3 mm/hour with every decade over the age of 30 years," but in temporal arteritis it is usually over 70 mm/hour. Since the arteritis is patchy, both circumferentially and longitudinally, a biopsy specimen from the temporal artery does not always show the typical multinucleated giant cells; but other changes, such as reduplication or fragmentation of the internal elastic lamina and round cell infiltration with intimal thickening, are more often seen. Treatment with steroids should be given at once-without waiting for the biopsy results-since recovery of vision is possible.'2 The initial treatment should be prednisone 15 mg four times daily, reducing as the ESR comes down to near normal limits, usually within three weeks. If the headache and other symptoms are not ameliorated within 48 hours the diagnosis should be reviewed. Treatment should be continued for a minimum of six months, the dosage being titrated against the clinical well-being of the patient and the ESR. Temporal arteritis is a self-limiting disease that lasts on average about a year,' but relapses have occurred with stopping treatment, even after several years. Cullen, J F, and Coleiro, J A, Survey of Ophthalmology, 1976, 20, 247. Huston, K A, et al, Annals of Internal Medicine, 1978, 88, 162. McLeod, D, British Journal of Ophthalmology, 1976, 60, 551. McLeod, D, and Kohner, E M, New England Journal of Medicine, 1977, 297, 1180. 5 McLeod, D, et al, British3ournal of Ophthalmology, 1978, 62, 591. 6 Winter, B J, Cryer, T H, and Hameroff, S B, Southern MedicallJournal, 1977, 70, 1479. 7Barricks, M E, et al, Brain, 1977, 100, 209. 8 Long, R G, Friedmann, A I, and James, D G, PoJstgraduate Medical 7ournal, 1976, 52, 689. 9Kansu, T, et al, Archives of Neurology, 1977, 34, 624. lo Gill, C R, Journal of the Kentucky Medical Association, 1977, 75, 483. 1 Hayes, G S, and Stinson, I N, Archives of Ophthalmology, 1976, 94, 939. 12 Parsons-Smith, B G, in Medical Ophthalmology, ed F Clifford Rose, chapter 23. London, Chapman and Hall, 1976. I 2
Early diagnosis is essential in temporal (giant-cell) arteritis to prevent blindness. The possibility must be considered in any patient over the age of 55 who presents with severe persistent headache, particularly if there are also systemic features such as malaise, anorexia, loss of weight, fever, night sweats-and especially pains in the limbs or joints that are worse in the morning and aggravated by exercise. Another valuable diagnostic pointer is pain in the masseter muscles on chewing. Examination will show temporal arteries that are tortuous, thickened, pulseless, and-most importanttender, so that the patient may say that combing her hair or wearing a hat has become impossible. Visual loss is the first complaint in nearly half of all cases': usually sudden, unilateral, permanent, and complete but less often gradual, bilateral, transient, and partial.2 The ophthalmoscopic findings are variable and depend on the vessels affected, but the most common pattern is ischaemic optic neuropathy due to posterior ciliary arteritis. In addition to narrowed vessels and a few circumpapillary haemorrhages, there may be oedema of the disc due to blocking of axoplasmic flow at the nerve head.3 This interference with axoplasmal transport is responsible for the transient cotton wool spots seen later on. A similar appearance develops in occlusion of the central retinal artery4 5; with complete occlusion the cotton wool spots accumulate at the optic nerve head, but with partial occlusion they occur at the junction of ischaemic and non-ischaemic retina, and are hence circumpapillary; they are not seen in the periphery. Owing to the pallor of the surrounding ischaemic retina a cherry red spot is seen in central retinal artery occlusion. The ocular vessels most often affected are the ophthalmic, central retinal, and short posterior ciliary arteries,6 but the long posterior ciliary arteries may also be damaged, resulting in the rare anterior segment ischaemia. Other uncommon ocular presentations include diplopia, which is often transient,7 and scleritis.8 The disease affects small and medium-sized vessels throughout the body and not just those of the head; the vessels most frequently found to be affected at necropsy are those containing elastic tissue, such as the extracranial vertebral arteries. The diagnosis is confirmed by finding a raised erythrocyte sedimentation rate (ESR), but this is not invariable.9 10 The
3 4
Coronary arteriography before aortic valve replacement The prognosis of patients with symptomatic aortic valve disease is poor. Without specific treatment the average life expectancy in those with aortic stenosis' is no more than three to four years. But replacement of the valve restores health and provides a good prospect of prolonged survival. The risks and results of surgery are largely determined by the state of the myocardium and its arterial supply, which cannot be decided on clinical grounds. Angina pectoris, for example, may be due to the valve disease alone, or to additional coronary artery disease.2 Conversely, coronary artery disease may be found in those without angina,3 4 albeit less commonly. Thus coronary artery disease can be identified only by coronary arteriography-a procedure that has a low risk of mortality or morbidity.5 6 Arteriography is clearly indicated if there is doubt about the severity of the aortic valve disease, but should it be routine ? In those centres where it has become a standard procedure the incidence of coronary artery disease in patients being assessed for aortic valve surgery has generally been found3 4 7to be between 20% and 3Oo%, but it may be8 as high as 48%. Coronary disease may need to be identified before replacement of the aortic valve for two reasons. Firstly, the operation
1444
may be made safer. In the 1593 operations for isolated replacement of the aortic valve in Britain in 1977 there were 135 hospital deaths-a mortality rate of 8-3%.9 This reflects surgical practice throughout the country, but there was considerable variation among centres: the rate in a good centre accepting all comers should be around 500.10 Most deaths are due to myocardial damage occurring during surgery, usually frank infarction.1' The standard method of protecting the myocardium during the operation has been to perfuse the coronary arteries with blood, and if the surgeon knows accurately the anatomy of the left main coronary artery he may be prevented from selectively cannulating one branch, thus causing underperfusion of the other.12 Nevertheless, coronary cannulation carries some risk, and local dissection of the arterial wall accounted for two of the deaths in the series quoted above." This and other reasons have persuaded cardiac surgeons to turn from coronary perfusion for minimising myocardial damage, and to use instead cardiac cooling with potassium-induced cardiac arrest. This method of myocardial preservation causes less damage'3 and is associated with a lower incidence of operative myocardial infarction'4-and there is no need for preoperative assessment of the coronary arteries. So the first justification for preoperative coronary arteriography is becoming obsolete. The second reason, however, is more cogent. Preoperative arteriography should help to improve the long-term results by identifying lesions in the coronary arteries that warrant bypass surgery. Replacement of the valve alone brings symptomatic relief in most patients with aortic valve disease,'5 as might be expected from a knowledge of the pathophysiology; but a few patients are left with persisting angina. Furthermore, patients with coexisting coronary artery disease do less well after surgery compared with those with normal coronary arteries.'6 So in suitable patients the combined operation would be desirable on both counts-were it not for the increased operative risk. This is difficult to measure. Some authors report no increased risk,'7 whereas others have found a threefold increase in the mortality rate.'8 In British units in 1977 the mortality rate for combined aortic valve replacement and coronary artery bypass grafting9 was 11.7%-higher than that for isolated aortic valve replacement, and likely to remain so because of the longer time taken for the operation. So we can draw no firm conclusions about the place of coronary arteriography before aortic valve replacement. Among the many unanswered questions are the functional importance of arteriographic narrowings in patients with aortic valve disease, the relative risks of isolated aortic valve surgery and the combined operation, and the long-term fate of vein grafts. For the time being cardiac centres have evolved their own policies: those using coronary cannulation rely on detailed knowledge of the anatomy of the arteries, while those using cardioplegic techniques undertake coronary arteriography in patients with angina and then in each individual decide the merits of bypass grafting. What does emerge from the confusion is the variation in hospital mortality rates among different centres in Britain, which (for elective aortic valve replacement) may be as low as 1%.19 Any hospital undertaking open heart surgery has a duty to make its results available to referring physicians so that they can take decisions about the surgical treatment of aortic valve disease on the basis of reliable
information. 1 2
Mitchell, A M, et al, American Heart_Journal, 1954, 48, 684. British MedicalJournal, 1977, 1, 597.
3 Swanton, R H, et al, British Heart_Journal, 1977, 39, 1347. 'Harris, C N, et al, British HeartJournal, 1975, 37, 656.
BRITISH MEDICAL JOURNAL
2 JUNE 1979
Pridie, R B, et al, British HeartJournal, 1976, 38, 1200. Lacy, J, et al, Annals of Thoracic Surgery, 1977, 23, 429. 7 Baxter, R H, et al, British Heart3Journal, 1978, 40, 918. 8 Linhart, J W, Journal of Thoracic and Cardiovascular Surgery, 1968, 55, 811. 9 United Kingdom Cardiac Surgical Register. London, Society of Thoracic and Cardiovascular Surgeons of Great Britain and Ireland, 1978. 10 Ross, J K, et al, British Medical_Journal, 1976, 2, 1485. 1 Sharratt, G P, Rees, P, and Conway, N, Journal of Thoracic and Cardiovascular Surgery, 1976, 71, 869. 12 Furlong, M B, et al,Journal of Thoracic and Cardiovascular Surgery, 1972, 63, 185. 13 Richardson, J V, et al, Circulation, 1979, 59, 75. 14 Ellis, R J, et al, Circulation, 1978, 58, suppl 1, 57. 15 Baker, C, and Somerville, J, British MedicalJournal, 1964, 1, 197. 16 Copeland, J G, et al, Journal of Thoracic and Cardiovascular Surgery, 1977, 74, 875. 17 McManus, Q, et al,3Journal of Thoracic and Cardiovascular Surgery, 1978, 75, 865. 18 Loop, F D, et al, Circulation, 1977, 55, 169. 19 Parker, D J, and Seabra-Gomes, R S, British HeartJ'ournal, 1974, 36, 1039. 5
6
Senile disciform macular degeneration The latest statistics' from the Department of Health show that macular lesions account for over 20% of the blind in the age group 65-74 and for over 45%0 in the over 75s. Presumably almost all of these persons have senile macular degeneration and some of them, in turn, have senile disciform macular degeneration. Through the ophthalmoscope, senile macular degeneration is seen as an area of mottling with small clumps of pigment, sometimes with yellow spots called drusen, centred roughly 21 disc diameters temporal to the optic disc. The area of degeneration extends over one or two disc diameters. The cause of senile macular degeneration is unknown, but its tendency sooner or later to become bilateral makes it a ,disabling condition: the central area of the patient's field of vision is progressively affected so that reading or recognition of faces gradually becomes impossible. There is no satisfactory treatment. Nevertheless, the sufferers are always encouraged to be told that the disease will not spread outwards beyond the central 10 degrees or so to blot out their peripheral field, which is so valuable for navigation, as in walking and crossing roads. Some patients with macular degeneration become blind suddenly as a result of the development of disciform lesions. Studies using intravenous fluorescein and retinal photography have shown that an important stage in the evolution of these lesions is the appearance of new small vessels beneath. the pigment epithelium of the macular area of the retina.2-9 The new vessels are fragile and may bleed, suddenly destroying the centre of the visual field: in such cases there is a large sheet of subretinal haemorrhage temporal to the optic disc. The ophthalmoscopic picture may be mistaken for a central or tributary retinal vein occlusion. The haemorrhage may spread into the retina and also sometimes into the vitreous, with the final lesion taking the form of an irregular lumpy plaque of fibrous tissue. At that stage little if any recovery can be expected and no treatment of that eye is possible. We have, however, accumulating evidence that argon laser photocoagulation can destroy the new subretinal vessels in their early stages and thereby prevent the disastrous haemorrhage, provided that the new vessels have not yet involved the foveola (the central portion of the fovea), where light coagulation would of itself destroy central vision.10-15 The new vessels may start at any point within the wider macular
