sign in sign up
<<
Home , Cone cell, Disc shedding, Eye, Foveola, Retina, Rod cell

Eye (1987) 1, 282-295

The Ageing : Physiology or Pathology

JOHN MARSHALL London

Summary The human'fetina is a unique component of the in that throughout life it is continuously exposed to optical radiation between 400 and 1400 nm. The physiology of the ageing retina and the regression in visual performance with age cannot therefore be studied in isolation, or discriminated, from the life long cumulative effects of radiant exposure. This paper describes the spectrum of age related changes in the retina as they merge imperceptibly between declining visual function and overt pathology.

Functional Loss ponents, the larger one being diminished Virtually every measure of visual function transmission by the ageing and a smaller demonstrates declining performance with one induced by neuronal changes in the age­ increasing age.! However, since all the com­ ing retina and cortex. ponents of the age in different ways it is Age related degradations in visual acuity5.6 essential when assessing changes in or and sensitivity7,8 have been docu­ neural function to allow for changes in the mented for many years9 and are especially stimulous incident upon the retina due to evident over the age of 60 years. 5 Recently the either perturbations in the optic media or use of sinusoidal grating patterns generated physiological processes in the anterior eye. by a interferometerlO has enabled the For example, in the older eye the retinal irra­ independent assessment of degradation of the diance achieved by a given source with a given quality by the optic media and the geometry will be less than that in a young eye effective loss in resolving power of the neuro­ because of a reduced pupillary diameter nal components in the .Il,12 The (senile ), and decreasing transmission conclusion from such studies is that the age and increasing fluorescence properties of the related deterioration in contrast sensitivity is and lens. The latter will also modify primarily caused by changes within the neuro­ the spectral irradiance as absorption and scat­ nal elements of the visual system. 12 ter in the ageing lens and cornea are more significant for shorter wavelengths.2 Ageing Populations The practical difficulties in relating a par­ Given the increasing socioeconomic impor­ ticular functional loss to a specific anatomical tance of rapid assimilation of visually site are slowly being resolved. For example communicated information any impairment McFarland and colleagues3 demonstrated a of vision severely handicaps the performance two times elevation of threshold in the dark of everyday tasks. As the handicap progresses adapted eye for every year over the age of 20. economic and social dependence of the Subsequently Gunkel and Gouras4 showed affected individual increases accordingly, that this elevation resulted from two com- Whilst a detailed analysis of population

Correspondence to: Professor John Marshall, Department of Clinical , Institute of Ophthal­ mology, Judd Street, London WCIH 9QS, THE AGEING RETINA 283 dynamics is beyond the scope of this article it such systems is phenomenal with entire is worth highlighting some worrying trends in populations changing every one to six days the industrialised world. In many western depending upon the tissue. Each day, some countries an increase in life expectancy 500,000,000,000 cells are shed and replaced coupled with a falling birth rate is giving rise to throughout the body, and in an average life­ ageing populations, for example, by the end time cell populations may renew more than of this decade 15.5 per cent of the population 12,000 times. In contrast cells in non-dividing of theUnited Kingdom will be over the age of systems are usually found in highly differen­ 65.13 Within this population there will be 2 tiated forms and do not undergo nuclear divi­ million more women than men, a finding of sion after their initial development in early significance as there is evidence that many embryonic life. senile eye conditions have an earlier onset and A special problem occurs when a rapidly a higher incidence in women.14 Perhaps more resynthesising cell system is trapped between worrying is the fact that the elderly constitute two non-dividing systems. In such a situation the major patient load for ophthalmic practi­ one or both of the non-dividing systems may tioners as there is a dramatic rise in the inci­ suffer secondary damage as a result of coping dence of ocular pathologies in this group. In a with the waste products generated by the trap­ recent study based on government sources the ped dividing system. figures for blindness in England and Wales were quoted as less than 10 per 100,000 in the age group 0-4 years and rising to 5,200 per Molecular Ageing 100,000 in the over 75 year 0lds.1s The figures At a molecular level both cell systems are whilst disturbing in themselves become much undergoing constant renewal and replace­ more worrying when extrapolated on the basis ment. Indeed it is this very process of continu­ of the Melton Mowbray study16 which would ous self synthesis that defines life (i.e. the suggest that blindness in the elderly is under­ maintenance of biological integrity). The only estimated, in government figures by at least exception to this process of constant molecu­ one third. Of this elderly blind patient group lar formation and degradation is the genetic approximately 20 per cent will have been reg­ material itself, DNA, which in non-dividing istered because of , but between 46 systems is perfectly stable. However even and 50 per cent will be the result of macular here if a lesion occurs in the double helix then disease. In essence senile macular degenera­ special enzymes detect it, remove the tions are now the leading cause of blindness in damaged part and attempt to replace it. This . the western world. unique process of molecular repair rather than replacement has only been described for Cellular Ageing DNA, and unlike the constant process of The ageing process is usually monitored by molecular replacement it is only initiated by reference to a single biologically independent the presence of damage in the double helix. variable, the passage of time.17 As a result it is Continued cell life is dependent on not often mistakenly considered as a solitary pro­ exceeding or overwhelming this repair pro­ cess. This is clearly not the case as at a cellular cess. DNA damage may result from a cell level even the most simplistic approach must being irradiated by high energy in recognise the differences between those cell short wavelength or ultra radiation systems undergoing cell division throughout and the repair process may slow down with life and those systems that do not. For clarity age. Both of these concepts will be discussed we can designate the former cells as 'Dividing' later. and the later to 'Non-dividing' categories. Molecular replacement is the cell's answer Dividing systems are usually relatively simple to the principle of entropy whereby all natural cells and are found in locations where cells are processes strive to dissipate energy and under constant attack from physical agents, thereby exist in an increasing state of dis­ such as surface cells of skin or cornea, or cells order. Any living cell is a highly ordered lining the gut. The rate of cell replacement in system and expends large amounts of energy 284 JOHN MARSHALL to retain its ordered structure by constantly oped inner segments. In all other locations the renewing it. presumptive photoreceptor cells are still The rates of molecular renewal vary dra­ undergoing mitotic division. For some as yet matically for example the is unexplained reason further development of renewed every 1 to 1.5 hours. IS In contrast the macular photoreceptor cells then seems to incorporated in rod disc membranes cease, and those in the rest of the retina last for two weeks19 whilst those constituting rapidly overtake them during the remaining the axoplasmic framework in a human retinal months in utero. At birth the macular photo­ ganglion cell may exist for 3 months or more. receptors are the most poorly developed, and In the latter case during their lifetime they do not reach full maturity until at least four to slowly migrate from their site of synthesis in six years post-partum.22 the cell body to their site of destruction at their in the optic tectum.20 Several The Adult Macula factors also influence the rates of molecular As ageing of the macula is the most significant renewal, including nutrition, temperature, clinical problem, discussion will concentrate and radiant exposure. The latter exerting its on this region. The term macula means patch, effect both in the establishment of diurnal and to some extent this conveys the clinician's rythms in biological organisms, and by induc­ dilemma in defining the boundaries of this ing photochemical changes in macro­ region. Ophthalmoscopically an individual molecular systems. Diurnal rythms operate at macula is considered to be an area centred on both a cellular and molecular level, and there the fovea throughout which there is an altered are many reports in the literature describing light reflexwhen compared with the surround­ cyclic fluctuation in mitotic activity and pro­ ing . The origins of this altered light tein synthesis according to the time of day or reflex are complex with contributions from an night.20 Photochemical processes always altered retinal thickness, densely pigmented create problems for cells as the reaction prod­ retinal and the presence ucts or intermediaries are often toxic, how­ of a yellow pigment, the macula lutea.23 The ever in most natural exposures the cell's extent of the ophthalmoscopic macula is defence mechanisms or renewal systems can highly variable between individuals, but com­ cope with the extra stress. In the case of monly occupies a 3 mm diameter zone, 4 mm unnatural or excessive exposures these temporal and 0.8 mm inferior to the optic defences may be overwhelmed and perma­ disc, being directly on the optic axis of the eye. nent damage may ensue. The anatomical macula may or may not lie within the same boundaries as the Embryology ophthalmoscopic region, as the former is The initial step in the ageing of any system is defined as that area of the retina where the its and for the retina number of layers of nuclei in the ganglion cell the process really begins in the firstfew weeks layer exceed two. of life. The macula is apparent by about 11 to At the centre of the macula is the fovea, a 13 weeks and can be clearly seen as a densely central concavity or pit formed by the radial pigmented area on transilluminating globes at displacement of the inner retinal layers during this age. The pigment epithelial cells within development.24 At the rim of the foveal pit the this region are taller but less wide than those retina is extremely thick (300 /lm) due to both in other locations.21 They retain this unique the number of displaced intermediary neu­ aspect ratio throughout life and it probably rones and also the large number of ganglion accounts for the density of their pigmentation cells that relate to the fovea The sloping walls i.e. by concentrating pigment granules/unit of the pit referred to as the foveal , area relative to the axis of incident light. The rapidly thin the retina to form the floor of the photo-receptor cells that overlay these macu­ pit, or the where the retina is at its lar pigment epithelial cells are also highly thinnest (150 /lm). Both the rim and floor of differentiated as at 13 weeks this is the only the fovea give rise to ophthalmoscopically area in which cone cells have already devel- visible reflexesin the young adult. The former THE AGEING RETINA 285 causes a brilliant, horizontally elliptical sur­ attachments to the face reflex called the foveal reflex, whilst the of the retina. 30 At about this time the brilliant latter gives rise to a minified or spot-like shimmering foveal and foveola reflexes of foveola reflex. youth are lost, giving evidence of an altered The foveola is the only rod free area of the topography of the retinal surface. This may retina and is also the centre of the foveal ava­ arise in part due to shallowing of the walls of scular zone. Retinal are present in the foveal pit resulting from an age related the inner retinal layers from about half-way loss of retinal neurones. A recent study on up the slopes of the clivus. It is generally con­ human optic demonstrated that a loss of sidered that both the thinning of the retina almost 50 per cent of retinal ganglion cells can and the absence of blood vessels contribute to be expected over a 70 year life span31 and it the high acuity attained by this area. should be remembered that 50 per cent of The foveal cones are highly specialised in a these cells serve the macula. number of ways. First, in the young, their inner and outer segments are of extremely Photoreceptor Cells small diameter 1.5 !-tm)thus facilitating a close The light-sensitive portions of the rods and packing density. Secondly because these cells cones consist of distributed have to enter into synaptic association with within a series of hollow 'coin like' mem­ horizontal cells and bipolars that have been branes. In rods the membranes or discs are radially displaced out of the foveola they have discrete units isolated from each other and the developed a long axonal-like component boundary membrane of the cell, with about called the inner connecting fibre. These fibres 1000 discs in each rod, In cones the disc mem­ run from the inner aspect of the nuclear region branes are all joined together, being continu­ to the cells synapses in the outer plexiform ous with the boundary membrane of the cell, layer. Collectively these fibres form the fibre The action of light upon these membranes layer of Henle, and it is within these elements leads to the production of a number of toxic that most of the macular pigment is products such as free radicals. located.25.26,27 Because of these long inner con­ necting fibres, foveal cones are the largest Rod Cells photoreceptor cells in the retina, some being Both rods and cones attempt to limit the over 500 !-tm long from tip of their outer seg­ damaging effects of ageing and light by con­ ment to their synaptic surface,25 Inner con­ stantly renewing their discs,19 Rods have a necting fibres are particularly well highly efficient process whereby 1-3 discs are demonstrated in isolated receptor cells (Fig, manufactured each hour and thus between 30 1). and 100 news discs are made each day. With The differences in retinal pigment epithelial successive disc production, units are pro­ cells in this region have already been men­ gressively displaced down the outer segment tioned, but it should also be stated that towards the pigment epithelium, Such is the Bruch's membrane and the underlying capacity of the system that the entire light are thicker in the posterior pole than sensitive portion of the is renewed in other regions of the eye and that the every two weeks. In order to prevent large choriocapillaries is more extensively fluctuations in rod length, old discs are developed2B. removed from the tips of the outer segment by the phagocytic action of the retinal pigment The Ageing Macula epithelium, This process seems to be initiated No single quality of the posterior pole of the by the onset oflight, i.e, rod tips are 'eaten' on eye remains constant during life, although awakening in the morning. However, it may individual age-related changes are of differing also be mediated by hormones, as a diurnal importance and portent.29 Commonly the col­ rhythm of is exhibited even in lapse of the vitreous gel in the fourth or fifth prolonged periods of . To date, evi­ decade of life is followed by a detachment of dence suggests that the underlying diurnal the posterior vitreous from its perimacular rhythms are initiated within the eye, and are 286 JOHN MARSHALL

Fig. 1. Scanning Electron Micrograph of Isolated Rod Cell. The various portions of this photoreceptor can be easily identified as follows: s: , i: inner connecting fibre, n: nucleus,!: outer connecting fibre, m: inner segment, c: , 0: outer segment The bar marker is 10 I-lm THE AGEING RETINA 287 not influenced by removal of the pineal mentosa sufferers, a drop out of photorecep­ gland.32 tor elements from the matrix is followed by an This rapid and continuous replacement of increase in diameter of those remaining. discs on an incremental basis, with each disc Typical figures for the diameter of cone inner being a discrete unit in total isolation from its segments available in the literature show a neighbours and the cell's boundary mem­ range of values from 1.2 to 4.5 [lmY We have brane, seems to confer adequate protection to shown that this distribution is age related, and the rod cells during the average human life­ that the rate of increase in cone diameter and time of 70 years. The result is that little evi­ presumable cell loss accelerates after the age dence can be found of loss of rod cells in the of 40 years (Fig. 2). ageing retina. Indeed so effective is the rod Light absorption in renewal system, that the only anomaly mani­ outer segments is directly related to the con­ fested with increasing age is a net increase in centration and orientation of molecules of the numbers of disc membranes.33 This visual and this in turn is dependent increase in discs is not accompanied by an upon the number and s.ize of photoreceptor increase in the axial length of the rod outer discs and their orientation. Both progressive segment and in consequence, in most indivi­ lamellar disorientation and drop-out duals over the age of 45 the rods develop would contribute to diminished light absorp­ convolutions in their outer segments. The tion by visual pigments within the fovea with increase in rod disc content infers that the rate increasing age. Such a finding has recently of disc formation does not slow significantly been reported in subjects between the ages 22 with age; on the contrary, it would either sug­ and 50 years by a group using the technique of gest a slight acceleration in disc production or to measure cone a decrease in the rate of phagocytosis; and as pigment density difference spectra. 38.39 will be seen later the latter is probably the It is also interesting to note that acquired mechanism. defects of colour vision in particular in blue­ discrimination are often associated with Cone Cells age or disease, the so-called Kollner's rule.40 Cones renew their discs at a much slower rate, Such acquired defects are particularly notice­ with the best evidence suggesting that the able in diabetics, and subsequent to periph­ whole system may take nine months to a year eral retinal ablation by xenon or argon to renew. 34 In contrast to rods, cone tips are photocoagula tors an acute increase in severity 'eaten' at night. The apparently poor renewal of the effect is noted41 and is thought to be rate of cone membranes may relate to the related to the massive doses of scattered intra­ findingthat cone 'discs' are not discrete, being ocular light. Whether the defects in blue per­ joined together in a continuum. This con­ ception result from a greater sensitivity of formation precludes individual disc replace­ short wavelength cones to ageing, light or ment on an incremental basis. Molecular disease, or whether they become apparent in renewal is a random process therefore in such this type of photoreceptor because there are a fluidsystem as the cone outer segment mem­ fewer of them in the cone population, is brane new molecules are just as likely to be unclear. replaced as old ones. Clearly this renewal mechanism is less efficient than that in rods Retinal Pigment Epithelium and may account for the age-related changes Once inside the retinal pigment epithelium, observed in cones. In essence a small and the groups of engulfed discs called phago­ approximately linear loss of cone membranes somes undergo lysis. Some of the breakdown is seen between birth and 40 years and then a products are recycled but many are voided rate change occurs and membrane loss is more into the choroidal blood supply via Bruch's rapid.35 This loss of membrane order is even­ membrane. The pigment epithelium is a non­ tually reflected in a net loss of cone cells over dividing system and with increasing age there the age of 40 and especially from the fovea.36 is a net loss of cells from this layer with a As has been seen in the of pig- resulting increase in size of those remaining. 288 JOHN MARSHALL

(/)S I I I I I I I I I Q) LITERATURE c: 0 () • 1.2-2.8 Polyak • 1.S-6 Rochon- �4 - - Duvigneaud > • 0 2 Henle LL. - 2.0-2.S Kuhnt 0 2.1-2.8 & - • - - Schafer �3 2.3-2.S Dimmer E OJ 2.S Wadsworth G) CIJ 2.S Greff .... 3 Muller - �2 r- 3 Merkel c: • 3 Krause - 0 • 3.1-3.6 Welcker .... Q) - Q) 1 r- - E co l5 I I I I I I J 1 I 10 20 30 40 SO 60 70 80 90 100 Age in Years

Fig. 2. Graph showing the increase in diameter of inner segments of foveal cones with increa�ing age. The literature data is from Polyack.37

Coupled with this reduction in available cells, that this differential increases with age.46 A both the lytic and voiding capabilities of those major difference between cells obtained from remaining seems to decline and result in the these two locations is that those from the mac­ accumulation of incompletely degraded pha­ ula contain more lipofuscin than those from gosomal particles called lipofuscin granules.42 the peripheryY In a further series of experi­ In the absence of disease, lipofuscin becomes ments Boulton and Marshall have shown that apparent within the pigment epithelium by if pigment epithelial cells of any age are arti­ the age of 10 years, and by the age of 40 years ficially fed lipofuscin granules in their culture occupies some 8 per cent of the cytoplasmic media, cell division and metabolic activities volume of the cells.43 By 80 years this figure are reduced as lipofuscin content increases, has risen to more than 20 per cent. As senes­ and that beyond a certain critical level the cence represents the accumulation of abnor­ cells die.4R From these studies it would appear mal or functionless molecules the build up of that the onset of senescence in the retinal lipofuscin within the pigment epithelium is a pigment epithelium is at least in part due to clear example of an ageing cell. the accumulation of lipofuscin which in turn Recently some interesting observations reflects the demands made upon it by the con­ have been obtained from studies on human stant phagocytic load of photoreceptor cell retinal pigment epithelial cells grown in tissue membranes. culture. First it is well established that the rate Unlike any other macrophage which has a of cell division in culture depends upon the transient existence and having fulfilled its role age of the donor from whom the cells were by ingesting its target material dies to be obtained, with fastest growth curves being dis­ replaced by a new cell, the same pigment epi­ played by cells from the young.44.45 The new thelial cell must continuously engulf some observation is that in any given individual, three hundred million million discs during a 70 cells derived from the macula grow less well year life span. At about the age of 40 years the than those from the retinal periphery, and pigment epithelial cells become too full with THE AGEING RETINA 289 these waste products and they attempt to rid choriocapillaris which themselves increase in themselves by pushing them out through both axial and lateral dimensions and effec­ Bruch's membrane. tively reduce the size of intervening capill­ aries. As virtually all of the nutrition of central Bruch's membrane and choroid cones derives from the choroid, this combina­ Bruch's membrane is an acellular membrane tion of a greater volume of intervening that is usually subdivided into five substrata, material together with greater separation of of which two are basement membranes, two elements of choriocapillaris undoubtedly con­ are collagenous and the remaining one is com­ tributes to a further decline in metabolic posed of elastin. A useful analogy is to con­ transfer capacity. sider the fibrouslayers as the wires in a garden There are also some indications that pertur­ sieve, however, in this case with increasing bations in molecular transport may be age, the fibrous components increase in selectively affected as biochemical studies numbers. Thus the sieve of Bruch's mem­ have shown that the proteoglycans in Bruch's brane assumes a progressively increased resis­ membrane change with age54 and in other tance to the passage of particles. This basement membranes where such changes increased resistance, coupled with an occur negatively charged fieldsare established , increased outflow of cellular detritus, results and these impede the passage of similarly in an accumulation of debris which in turn charged macromolecules. 55 A further con­ tends to displace the overlying pigment epi­ sideration in relation to retinal fluid control is thelial cells. Over the age of 40 years, focal that the net flow of water is from retina to aggregations of such sub-pigment epithelial choroid, with the pump residing in the pig­ debris are observed and these excrescences on ment epithelium. If the build up of mem­ Bruch's membrane are called Drusen.49 branous debris within Bruch's membrane with As well as the alterations induced in age is predominantly lipid then a hydrophobic Bruch's membrane by phagocytic problems of barrier may be created which would preferen­ the retinal pigment epithelium other ageing tially impede this outward movement of water changes occur which are also thought to have from the retina. 56,57 functionafsignificance, and of these the most prominent is the accumulation of a fine granu­ The pathology of the Ageing Macula lar deposit on its innermost margin, termed by All of the previously described age related Sarks the 'Basal Linear Deposit'. 49 This changes in the physiology of the macula may deposit is a thickened and chemically modi­ ultimately contribute to a pathological state. ned basement membrane of the pigment epi­ Some of the consequences of senescence are thelium, In association with these basement described below. membrane changes the basal border of the epithelial cells undergo a reduction in the Drusen complexity of its convolutions, a finding that All eyes over the age of 40 have the precursors would suggest a progressive decline in active of Drusen within Bruch's membranes in the transport with age. peripheral retina. In such locations, Drusen Within the fibrous layers of Bruch's mem­ do not seem to lead to sight-threatening brane changes are also apparent. 52 The, col­ sequelae.50 However, if they occur in the lagen fibres become more numerous with age central retina they are identifiedas risk factors and there is an increasing incidence of fibres associated with more significant retinal path- displaying an atypical banding periodicity. 010gy,51 If, as is presently postulated, Drus�n The fibre numbers also increase in the elastic result from the age related inability of the layer, but here calcification is the most worry­ pigment epithelial cells to fully digest rod ing feature as the resultant changes in elas­ outer segment debris then the macular distri­ ticity renders Bruch's membrane more bution of Drusen is easy to explain. If the brittle.53 Such changes extend throughout the geographic distribution of rods determined by thickness of Bruch's membrane to involve the Osterberg52 is compared with the distribution intercapillary pillars of the adjacent of lipofuscin determined by Wing and Col- 290 JOHN MARSHALL

TEMPORAL NASAL

"TI r­ 160 LIPOFUSCIN C RODS -- 18YEARS 1.8 o II I m C') CONES PREMATURE-12 MONTHS ••• I (J) o 140 1.6 (") 61YEARS-88YEARS ••• II m II z II 1.4 (") 120 '. II m (J) • w II • o z II 1.2 "TI • o 100 I I r­ u • • "1J • II a: I I '" • 1.0 o o H '" • "TI 80 • o C (J) • • (J) o � 0.8 (") o t­ a: Il. • Z 60 o • u. • 0.6 J> o • • � "1J • • ci 40 (J) z • I 0.4 • I • )( • • •• . • • •• . 20 • . • • L , o . . . . . 0.2 . . \ . . . I . " . · ------. (Jl I. · · - ---- : .- ��� . . OO�_.�--L---�--L---L---�--4_-L�--�--�--+_--��UL��O.O ORA EQUATOR FOVEA EQUATOR SERRATA

Fig. 3. Graph to show the relative topographic distributions of rods and cones in relation t9 the distribution of lipofuscin. The rod and cone distribution is taken from the workd of Osterberg58 whilst that of lipofuscin is taken from the work of Wing and colleagues. 47 It should be noted that the maximal distribution of lipofuscin coincides with the maximal distribution of rods.

leagues47 then the two can be seen to be Senile coincident (Fig. 3). Senile macular degeneration (SMD) again results from the metabolic disadvantage resulting from the abnormal accumulation of debris between the pigment epithelium and Retinal pigment epithelial detachment the choroidal blood supply. In this case the Here, in relation to, or exacerbated by, macular and foveal photoreceptors atrophy. Drusen or sometimes as a result of other The factors controlling the underlying aetiol­ causes such as a thickened basal linear ogy are not known but several authorities sug­ deposit, the pigment epithelium detaches gest a strong familial or genetic influence.51.59 from Bruch's membrane. If a hydrophobic Senile macular degeneration is the com­ barrier is present within Bruch's membrane57 monest form of macular disease and is fre­ and if the predominant flow of fluid is from quently bilateral.16 Whilst �pidemiological retina to choroid then fluid will pool between studies are poor, a recent survey in the USA the pigment epithelium and Bruch's mem­ gave a prevalence for both eyes of 4 per cent brane. With increasing time the pigment epi­ between 65-74 years and 17 per cent between thelium will literally pump itself off Bruch's 75-85 years, with a 22 per cent prevalence for membrane.56 If sufficient fluid accumulates single eyes in the older group.60 In the United under the area of detachment, or if the pig­ Kingdom, in the over-70 year age group, ment epithelial barrier changes, or if the epi­ senile macular degeneration is responsible for thelium , the overlying photoreceptors as many as 47 per cent of blind registrations61 may die as a result of the induced disturbances and 63 per cent of low vision cases.62 In all in homeostasis. studies, except those in Australia63 women THE AGEING RETINA 291 show a higher prevalence than men with an stand and it is difficulteven to relate the site of earlier onset. Given the ageing populations of primary damage to a particular cell type, as the western world and the longevity of several apparently conflicting pieces of women, there will be a marked increase in the empirical evidence have to be considered. numbers of people suffering from senile mac­ First it should be explained that, as with any ular degeneration. Unless treatment or pro­ biological reaction, a spectrum of light­ phylatic measures improve, we must prepare induced changes can be expected and that for the social and economic consequences of particular changes within such a spectrum will an increasing blind population. be favoured by particular exposure regimes. Secondly, because of lack of knowledge of the Disciform lesions total spectrum, present accounts may tend to There is a further complication in senile macu­ polarise and isolate damage mechanisms in a lar degeneration in which first macrophages way which may subsequently prove and then choroidal blood vessels begin to unjustified. At present, experimental data invade Bruch's membrane. These vessels may exist for four different types of exposure regi­ penetrate the pigment epithelium and thereby mes and these seem to indicate the existence circumvent its blood retinal barrier properties of two different mechanisms. The exposure and they may also rupture and bleed with regimes are in order of increasing irradiance: diastrous consequences for vision.64,65 They (a) prolonged exposure to commercial or also have an unfortunate behavioural prop­ domestic levels of fluroescent lighting erty in that whatever their site of origin they (several hours to several days).69.70 tend to grow towards the fovea where their (b) repetitive exposure to narrow band leaky properties cause rapid atrophy of the monochromatic (several hours/ photoreceptor cells. Currently laser treat­ day for some days).71 ment is only really effective if the patients are (c) low level non-thermal exposure to diagnosed very early in the of the (10 seconds to several minutes)72 lesion and recurrences are such that one esti­ and mate is that only between 30�00 people are (d) relatively short exposures to clinical kept off the blind registration per year for all light sources (several minutes).73 the treatment given for SMD. Regimes (a) and (b) result in damage that could be explained by absorption of light by The role of light in relation to retinal ageing visual pigments in the photoreceptor cells; The human retina is normally subjected to whilst (c) and (d) seem to be dependent upon irradiances below lQ-4 Wcm-2 and above this absorption within the pigment epithelium. level the exposure duration is limited by blink­ Photoreceptor cells. The initial indication ing. We do not know if the changes in our light that moderate levels of illumination from arti­ environment over the past 50 years are affect­ ficial light sources could result in retinal ing retinal performance or disease processes; damage came in the mid-sixties when it was we do know that artificial light regimes have shown that if rats were kept under constant dramatic effects on the retinae of experi­ illumination from fluorescent lamps, their mental animals66 and that in commerce and photoreceptor cells degenerated.74 The origi­ industry, recommended levels of illumination nal work of Noell and others was criticised are constantly being revised upwards.67,68 We because of the use of rats, a nocturnal also know that in the clinical environment we with a predominantly rod retina. However, have a worst-case situation with extremely when such studies were extended to diurnal bright lights and no functional physiological species a further surprising finding was made protective mechanisms, i.e. the blink reflex in that cones were affected at irradiances and aversion responses are commonly over­ lower than those required to damage rods.70 come by anaesthesia and pupillary closure by For example in the monkey, 12 h exposures . with retinal irradiances between 195 and 361 The changes in the retina induced by exces­ Wcm-2 damaged cones, whilst irradiances of sive exposure to light are complex to under- 361-615 Wcm-2 were required to produce rod 292 JOHN MARSHALL damage.7o These studies also showed that the laser light, the so called 'Blue Light Hazard'.77 fovea was apparently the most sensitive area; It has also enabled the relative retinal hazard however, it should be remembered that in of any light source to be more accurately this region of retina is said to recline assessed, as we now know that in addition to the highest irradiance. This rod-cone differen­ the total retinal irradiance achieved by a tial is almost certainly not a difference in source, the relative spectral emission also has threshold to primary damage, but probably to be considered in relation to Blue Light results from the different repair capacity of Hazard weighting functions.78 In this context the two cell types, i.e. rods and cones are it is of interest that the absorption spectra of probably damaged to the same extent but rods the ageing lens and that of the macular pig­ are better able to repair themselves and sur­ ment seem beautifully designed to filter out vive. Two further points should be empha­ the Blue Light Hazard before incident radia­ sised; first all these experiments were carried tion reaches the light-sensitive portions of the out on young healthy and effects on photoreceptor cells.79 Cataract surgeons the cones of age-compromised retinae may be should think very carefully of the possible far more dramatic. Secondly, light damage problems of suddenly exposing an age-com­ from fluorescent lamps always becomes more promised retina to an influxof potentially haz­ severe when exposure periods are sufficiently ardous wavelengths by removing an aged lens long to eliminate the normal diurnal cycle of full of blue light absorbing yellow pigment, light and dark.75 This may have interesting and should perhaps redress the consequences implications for succeeding generations as flu­ of their actions by the use of a suitably tinted orescent lights now gives us unlimited access artificial lens. 80 The importance of the absorp­ to daylight levels of illumination. tion properties of the lens in particular in rela­ In a second type of experiment, exposing tion to attenuating UVA, have been well monkeys to repetitive flashes of light from demonstrated by Ham and his colleagues81 monochromatic sources, Harwerth and Sper­ who demonstrated that in aphakic monkeys ling71 showed that cone loss was most easily "blue light" like retinal lesions could be elicited by exposure to blue light; a finding induced by UVA at energy levels six times which perhaps explains Kollner's rule? Thus lower between 325 and 350 nm than at the far then, light-induced damage has been blue light peak of 441 nm. described in photoreceptor cells, occurring in The current debate concerning whether or cones at lower irradiances than rods, and in not intraocular (I OLs) should or should short wavelength exposure at lower irra­ not contain UVA blocking filters is confusing diances or shorter exposure durations than for to the casual observer and much of the evi­ long wavelengths. dence quoted by the respective protagonists is Pigment epithelium. The second group of ambiguous. The concern of the antifiltration results that are attributed to photochemical group is 'do we have a treatment for a con­ mechanisms relate to relatively higher retinal dition that doesn't exist, and for which the irradiances delivered over shorter time spans. patient pays a commercial premium'. This These studies show that for supra-threshold group quotes the conflicting data on the inci­ exposures only, in the 10 s-2h exposure dence of cystoid macular oedema (CME) domain, the initial changes occur in the pig­ after in groups of patients ment epithelium. Detailed studies with lasers with and without UVA blockers. CME is how­ demonstrated a clear relationship between ever not the critical pathology in relation to thresholds for such damage and wavelength, monitoring the performance of IOLs. CME is with again lower irradiances being required to a relatively acute condition with multifactorial produce damage with short wavelength origins most of which relate to surgery. These radiation.72,76 can be summarised as factors due to vitreous Ham's data has resulted in most national manipulation, secondary inflammation, fluid and international bodies concerned with laser mobility from anterior to posterior , and safety incorporating wavelength-related light exposure during surgery, If light is an restrictions on prolonged viewing of low-level important cofactor then the exposure to the THE AGEING RETINA 293 microscope lamp during surgery is critical as and Refraction of the Eye. 1864; London: New this may initiate a subthreshold blue light or Sydenham Soc. 10 Campbell FW, Green DG: Optical and retinal Ham type lesion to the retinal pigment epi­ factors affecting visual resolution. 1. Physiol. thelium. A compromised pigment epithelium 1965; 181: 576-93. II with disturbed pump and barrier properties McGrath D, Morrison JD: The effects of ageing on could well ultimately result in accumulation of the optical quality of the . 1. Physiol. fluidin the retina and CME. To prevent CME (Lond.) 1983; 343: 81-2. filtration of the microscope lamp is probably 12 Morrison JD, McGrath C: Assessment of the optical contribution to the age-related deterioration in more important than a UVA absorber in the vision. Quart. 1. Exper. Physiol. 1985; 70: 249- IOL. In contrast, SMD is a chronic condition 69. rarely seen in eyes in which are pres­ 130PCS. Office of Population Censuses and Surveys ent. In this case subsequent to surgery the London, St. Catherine's House, 10 Kingsway, retina is going to be exposed to high levels of London, WC2B 6JP, U.K. 1983. 14 Knave B, Tengroth B, Voss M: Age and sex distri­ blue light and in the absence of a filter, UVA. bution of some retinal macular diseases: senile This sudden increase in short wavelength radi­ and presenile macular degeneration and central ation may well accelerate photoreceptor cell serous retinitis. Acta Ophthalmol. (In Press). damage due to photochemical processes of 15 Royal National Institute for the Blind: Initial Demo­ graphic Study. Shankland Cox, London, 1985. the non-Ham type. For example it has 16 Gibson JM, Rosenthal AR, Lavery J: A study of the recently been reported from the Wilmer Insti­ prevalence of in the elderly in an tute (ARVO 1986) that in a group of monkeys English Community. Trans. Ophthalmol. Soc. in which non UV absorbing IOLs had been U.K. 1985; 104: 196-203. inserted in one eye, there was a three times 17 Mainster M: Light and macular degeneration: A biophysical and clinical perspective. Eye. 1987; 1: higher incidence of macular drusen in the 304. operated eyes. In this case a UVA filter in the 18 Macri FJ, Dixon RL, Rail DP: Aqueous humor turn­ IOL may well have delayed the onset of senile over rates in the . Invest. Ophthalmol. 1965; 4: changes in the macula or at the very least 927-34. 19 Young RW: Visual cells and the concept of renewal. offered an element of protection. The use of Invest. Ophthalmol. Vis. Sci. 1976; 15: 700-25. the UVA blocking IOLs is relatively new, but 20 Young RW: The Bowman Lecture. Biological over the next few years studies on these Renewal. Applications to the eye. Trans. patients should include the pattern and inci­ Ophthalmol. Soc. u.K. 1982; 102: 42-75. dence of post operative SMD rather than 21 Streeten BW: Development of the human retinal pigment epithelium and the posterior segment. CME as a guide to their usefulness. Arch. Ophthalmol 1969; 81: 383-94. 22 Abramov I, Gordon J, Hendrickson A, Hainline L, References Dobson V, LaBossiere E: The retina of the new­ I Kline D: Aging and the spatiotemporal discrimina­ born human infant. Science 1982; 217: 265-7. tion performance of the visual system. Eye. 1987; 23 Nussbaum n, Pruett RC, Delori FC: Historic per­ 1: 323. spectives. Macular yellow pigment. The first 200 2 Weale RA: Senescent vision: Is it all the fault of the years. Retina 1981; 1: 296-310. lens. Eye. 1987; 1: 217. 24 Hendrickson AE, Yuode\is C: The morphological 3 McFarland RA, DorneyRC, Warren AB, Ward DC: development of the human fovea. Ophthal­ Dark as a function of age. I. A statis­ mology. 1984; 91: 603-12. tical analysis. 1. Gerontal. 1960; 15: 149-54. 25 Marshall J, Hamilton AM, Bird AC: Histopathol­ 4 Gunkel RD, Gouras P: Changes in scotopic visibility ogy of ruby and argon laser lesion in monkey and thresholds with age. Arch. Opthalmol. 1963; 69: Human Retina: A comparative study. Br. 1. 4-9. aphthalmol. 1975; 59: 610-30. 5 Slataper FJ: Age norms of refraction and vision. 26 Snodderly DM, Brown PK, Delori FC, Auran JD: Arch. Ophthalmol. 1950; 43: 466-81. The Macular Pigment. I. Absorbance Spectra, 6 Weale RA: Senile changes in . Trans. Localisation, and Discrimination from other Ophthalmol. Soc. u.K. 1975; 95: 36-8. yellow pigments in . Invest. 7 Arden GB, Jacobson JJ: A simple grating test for aphthalmol. Vis. Sci. 1984; 25: 660-73. contrast sensitivity: preliminary results indicate 27 Snodderly DM, Auran JD, Delori FC: The Macular values in screening for . Invest. Pigment. II. Spatial Distribution in Primate Reti­ Ophthalmol. and Vis. Sci. 1978; 17: 23-32. nas. Invest. aphthalmol. Vis. Sci. 1984; 25: 674- 8 Skalka HW: Effect of age on Arden grating acuity. 85. Br. 1. Ophthalmol. 1980; 64: 21-3. 28 Gass JOM: Pathogenesis of disciform detachment of 9 Donders FC: On the Anomalies of the neuroepithelium. 3. Senile disciform macular 294 JOHN MARSHALL

degeneration. Am. 1. Ophthalmol. 1967; 63: 617- Human Retinal Epithelium in vivo and in vitro. 44. Monogr, devl. Bioi. 1984; 17: 80--93. 29 Marmor MF: Aging and the Retina. In: Aging and 47 Wing GL, Blanchard GC, Weiter JL: The topogra­ Human Visual Function. R. Sekuler, D. Kline, J. phy and age relationship of lipofuscin con­ Dismuths (eds.) New York: Alan R. Liss. Inc. centration in the retinal pigment epithelium. 1982; 59-78. Invest. Ophthal. Vis. Sci. 1978; 17: 601-7. 30 Sebag J, Balazs EA: Pathogenesis of cystoid macular 48 Boulton ME, Marshall J: The effects of increasing edema: An anatomic consideration of vitreoreti­ numbers of phagocytic inclusions on human reti­ nal adhesions. Survy Ophthalmol. 1984; 28: nal pigment epithelial cells in culture: A model 493-8. for ageing. Br. 1. Ophthalmol. 1986; 70: 808-15. 31 Balazs AG, Rootman J, Drance SM, Schulzer M, 49 Sarks SH: Ageing and degeneration in the macular Douglas GR: The effect of age on the nerve fiber region: A clinico-pathological study. Br. 1. population of the human . Am. 1. Ophthalmol. 1976; 60: 324-41. Ophthalmol. 1984; 97: 760--6. 50 Foos RY, Trese MT: Chorioretinal junction: Vas­ 32 Flannery JG, Fisher SK: Circadian disc shedding in cularization of Bruch's Membrane in peripheral Xenopus retina in vitro. Invest. Ophthal. Vis. Sci. fundus. Arch. Ophthalmol. 1982; 100: 1492-503. 1984; 25: 229-32. 51 Gass JDM: Drusen disciform macular detachment 33 Marshall J, Grindle CFJ, Ansell PL, Borwein B: and degeneration. Arch. Ophthalmol. 1973; 90: Convolutions in human rods: An ageing process. 208-17. Br. l. Ophthalmol. 1979; 62: 181-7. 52 Grindle CFJ, Marshall J: Ageing changes in Bruch's 34 Foulds W: Factors influencingvisual recovery in reti­ membrane and their functional implications. nal detachment surgery. Trans. Ophthalmol. Trans. Ophthalmol. Soc. U.K. 1978; 98: 172-5. Soc. u.K. 1980; 100: 72-7. 53 Loffler KU, Lee WR: Basal linear deposit in the 35 Weale RA: The eye and ageing. Interdisipl. Topics human macula. Graefe's Arch. Clin. Exp. Geront. 1978; 13: 1-13. Ophthalmol; 1986, 224: 502-6. 36 Gartner S, Henkind P: Ageing and degeneration of 54 Hewitt T, Newsome D: Altered synthesis of Bruch's the human macula. 1. and membrane proteoglycans associated with domi­ photoreceptors. Br. 1. Ophthalmol. 1981; 65: nant retinis pigmentosa. Current Eye Res., 1985; 23-8. 4: 169-74. 37 Polyak SL: The Retina. University of Chicago Press. 55 Kanwar YS, Farquhar MG: Anionic sites in the 1941. glomerular basement membrane: In vivo and in 38 Kilbride PE, Read JS, Fishman GA, Fishman M: vitro localization to the laminae rarae by catronic Determination of human cone pigment density probes. J. Cell Bioi. 1979; 81: 137-53. difference spectra in spatially resolved regions of 56 Bird AC, Marshall J: Retinal pigment epithelial the fovea. Vision Res. 1983; 23: 1341-50. detachments in the elderly. Trans. Ophthalmol. 39 Kilbride PE, Hutman LP. Read JS, Fishman M: The Soc. U. K. 1986; 105: 674-82. ageing human eye and cone pigment density 57 Harper CA, Marshall J: The deposition of Lipids in difference in the fovea. Invest. Ophthalmol. Vis. Bruch's membrane and the implications for age Sci. 1984; 24: 198. related retinal pathologies. Br. 1. Ophtha/mol. 40 Kolner M: Die Storungen des Furbensinnes ihre 1987 (in press). Klinische Bedeutung und ihre Diagnose, Karger, 5R Osterberg GA: Topography of the layer of rods and Berlin 1912. cones in the human retina. Acta. Ophthalmol. 41 Birch l, Hamilton AM: Xenon arc and argon laser 1935; 13: Supp!. 6; 1-102. photocoagulation in the treatment of diabetic 59 Hyman LG, Lilienfeld AM, Ferris FL, Fine S: Senile disc neovascularization. 2. Effect on Colour macular degeneration: A case control study. Am. Vision. Trans. Ophthalmol. Soc. u.K. 1981; 101: J. Eipdemiol. 1983; 118: 213-27. 93-9. 6() Khan MA, Leibowitz HM, Ganley JP, Kini MM, 42 Feeney L: The phago lysosomal system of the pig­ Colton T, Nickerson RB, Dawber TR: The Fra­ ment epithelium. A key to retinal disease. Invest. mingham eye study. I. Outline and major preva­ Ophthalmol. Vis. Sci. 1973; 12: 635-8. lence findings. Am. 1. Expidem. 1977; 106: 43 Feeney-Burns L, Hilderbrand ES, Eldridge S: 17-32. Aging Human RPE: Morphometric Analysis of 61 HMSO. Blindness and partial sight in England 1969- Macular, Equatorial and Peripheral Cells. Invest. 1976. Reports on Public Health and Medical sub­ Ophthalmol. Vis. Sci. 1984; 25: 195-200. jects No. 129. HMSO London 1979. 44 Boulton ME, Marshall J, Mellerio J: Human retinal 62 Robbins HG: Low vision care for the over 80's. pigment epithelial cells in tissue culture: A means Aust. J. Optom. 1981; 64: 243-51. of studying inherited retinal diseases. 1982; 1816 63 Robbins HG: The low vision patient of tomorrow. March of Dimes Foundation. Aust. J. Optom. 1978; 61: 54-64. 45 Flood MT, Gouras P, Kjeldbye H: Growth charac­ 64 Lee WR: Degenerative and other disorders of the teristics and ultrastructure of human retinal pig­ retina and choroid. In Garner, A. and Klint­ ment epithelium in vitro. Invest. Ophthalmol. worth, G. K. (Eds.) Pathology of Ocular Disease: Vis. Sci. 1980; 19: 1309-20. A Dynamic Approach 1982; 1321-1351. Marcel 46 Flood MT, Haley lE, Gouras P: Cellular Aging of Dekker, New York. THE AGEING RETINA 295

65 Eagle RC: Mechanisms of . Ophthal­ Arnott, EJ (Eds.) 1983; Chap. 19: 182-207. C. V. mology 1984; 91: 613--26. Moseby, St. Louis. 66 Lanum J: The damaging effects of light on the retina. 74 Noell WK, Walker VS, Kang BS, Berman S: Retinal Empirical findings, theoretical and practical damage by light in rats. Invest. Ophthalmol. Vis. implications. Surv. Ophthalmol. 1978; 22: 221- Sci. 1966; 5: 450-73. 49. 75 Holiyfield JG, Basinger SF: RNA Metabolism in the 67 IES Code: Recommendations for good interior retina in relation to cyclic lighting. Vision Res. lighting. Illuminating Engineering Society, 1980; 20: 1151-5. London 1961. 76 Lawill T, Crockett S, Currier G: Retinal damage 68 IES Code: Recommendations for lighting buildings. secondary to chronic light exposure. Documenta Illuminating Engineering Society, London 1968. Ophthalmol. 1977; 44: 379-402. 69 Marshall J, Mellerio J, Palmer DA: Damage to 77 British Standards Institute: Radiation safety of laser pigeon retinae by moderate illumination from products and equipment manufacturing require­ fluorescent lamps. Exp. Eye Res. 1972; 14: 164-9. ments, user's guide and classification, BS.4803 0 7 Sykes SM, Robinson WG, Waxler M, Kuwabara T: 1983. Damage to the monkey retina by broad-spectrum 78 Sliney D, Wolbarsht M: Safety with lasers and other fluorescent light. Invest. Ophthalmol. Vis. Sci. optical sources. Plenum Press, New York. 1980. 1981; 20: 425-34. 79 Marshall J: Radiation and the Ageing Eye. Ophthal. 71 Hawerth RS, Sperling HG: Prolonged colour blind­ Physiol. Opt. 1984; 4: 1-23. 0 ness induced by intense spectral lights in Rhesus 8 Mainster MA: Spectral transmittance of intraocular monkeys. Science N.Y. 1971; 174: 520-3. lenses and retinal damage from intense light 72 Ham WT, Mueller HA, Sliney DH: Retinal sen­ sources. Am. 1. Ophthalmol. 1978; 85: 167-70. 81 sitivity to short wavelength light. 1976; Ham WT, Mueller HA, Ruffob JJ: Action spectrum 260: 153--5. for retinal from near- radiation 73 Marshall J: Light damage and the practice of oph­ in the aphakic monkey. Am. 1. Ophthalmol. thalmology. In Rosen, ES, Maining, WM and 1982; 93: 299-306.