Eye (1991) 5, 265-278

THE MONTGOMERY LECTURE

Some Factors Which Affect the Visual Outcome of Corneal Transplantation

D. 1. COSTER Adelaide, Australia

Summary Corneal transplantation, which began in Dublin early in the nineteenth century, is now widely practised. Most grafts are done for visual reasons, some to manage cor­ neal destructive disease. The factors associated with allograft rejection are well known, but a clear graft does not mean that the transplantation procedure has been successful. Successful corneal grafts enable patients to improve their life-style. To achieve this, the graft must be transparent and free of optical aberrations, the eye must be capable of achieving good vision, and the patient must have a life-style which would benefit from a successful functioning graft. A study of patients who have had corneal grafts identified the presence of a functioning graft, the lack of need for a contact lens, and the achievement of visual acuity in the operated eye which exceeds that of the contralateral eye, as requirements for patient satisfaction.

In accepting the invitation to deliver the 1990 lar of the 5 major codes played in Australia, Montgomery lecture, I am aware of the great and the origins of the game go back to the honour bestowed upon me. The list of pre­ 1850s in Victoria, when the Irish population of vious lecturers contains the names of famous the colony was the largest outside Ireland. 1 men who have made great contributions to From humble beginnings in the paddocks ophthalmology over many years. My only dis­ around Melbourne, the game has grown to its tinguishing characteristic is that I come from dominant position in Australian sport, draw­ further away from Ireland than anybody else ing crowds of 100,000 people to major games. who has had the honour of delivering this Although Australian Rules football is slightly prestigious lecture. Although Australia is the different from the Gaelic football played in most remote continent, this does not mean Ireland, the two countries play each other that Australian society has escaped the influ­ from time to time. ence of Irish culture, science and medicine. Corneal transplantation had its beginnings Two of my major interests, corneal trans­ in Ireland. In the late 18th century, there was plantation and Australian Rules Football, considerable interest in the prospect of cor­ had their beginnings in Ireland. neal transplantation overcoming the ravages Australian Rules Football is the most popu- of Egyptian ophthalmia. This serious eye

Correspondence to: Professor D. J. Coster, Department of Ophthalmology, Flinders Medical Centre, Bedford Park, South Australia 5042. 266 D. J. COSTER disease, which produced a great deal of blind­ place, and 10 days after the operation the ani­ ness and disability as a consequence of cor­ mal gave unequivocal signs of vision. The neal inflammation, was brought back to upper part of the transplant remaining per­ Europe from Egypt at the end of the 18th cen­ fectly transparent." tury. It led to the development of ophthal­ On his return to Dublin, Bigger carried out mology as a separate branch of surgery and a number of experiments, and reported the created a great deal of interest in the potential results of corneal transplantation in 18 rab­ of corneal transplantation, even at a public bits. These experiments included various level. Erasmus Darwin, Charles Darwin's combinations of allografts and autografts. Of grandfather, was the first in the English­ the 18 grafts he reported, 16 recovered imper­ speaking world to suggest a strategy for cor­ fect vision. He almost certainly observed the neal transplantation. In 1797 he wrote, corneal allograft response, although of course "Could not a small piece of the cornea be cut he did not understand its significance."O n the out by a kind of trephine, about the size of a 30th day, violent inflammation occurred in thick bristle, or a small crow quill, and would it , one of the rabbits, without any evident cause, not heal with a transparent scar?, 2 and terminated in a copious deposition of As Darwin suggested, there were two puriform lymph in the anterior chamber. At important questions about the biology of cor­ the end of 10 days it subsided." neal transplantation that needed to be Bigger recognised the potential importance answered. Firstly, would a graft heal, and of corneal transplantation and correctly pre­ secondly would the resultant scar be dicted that the technical aspects of surgery, transparent? the main thrust of his work, would be easier in Amongst the firstpractical contributions to humans because the eye was larger and the the science of corneal transplantation were subject was likely to be more co-operative.He those of Samuel Bigger, an Irish ophthal­ was also concerned about the source of donor mologist.3 In 1831 he travelled to Germany, where there was already some interest in the material. He suggested the cornea of the pig subject. In particular, Reisinger had com­ was likely to be the most useful, and his advice menced experiments with rabbits after was taken up by others interested in the sub­ becoming interested in the early skin grafting ject with, at least with hindsight, predictable work of Astley Cooper in London. Reisinger results. seemed to be more interested in the healing Bigger was also concerned about the use of process than the visual result. He reported a human donor tissue. "In a spirit of just and partially successful graft in the Bavarian humane feeling, he deprecates the removal of Annals of 1824. This partial success in one the cornea from the human eye, even when animal was all that was recorded, despite permitted for gain by the possessor; but thinks experiments being carried out in animals in a that a person affected with incurable amauro­ number of centres at that time. After his visit sis might be prevailed upon to part with his to Germany,' Bigger visited Egypt, " ... a pellucid cornea, which might be replaced with country noted for the prevalence of destruc­ one taken from some inferior animal." In his tive ophthalmia." Somehow, in 1835 he conclusion, Bigger attempted to predict the became " ... a prisoner with a nomadic tribe patients who were most likely to benefitfrom of Arabs, about 12 or 14 days journey from corneal transplantation. He was "of the opin­ Grand Cairo." It was under these circum­ ion that cases of blindness caused by small­ stances that he carried out the first corneal pox, ulcers on the cornea, and ophthalmia not graft to result in improved vision. affecting the deeper structures of the eye, "The subject of the operation was a pet would be the most favourable for operation." gazelle, who had lost one eye from inflam­ In his paper of 1837, he concluded, with a plea mation and the power of seeing with the to his colleagues " ... imploring hospital sur­ other, from a wound to the cornea. The geons to give the matter their attentive con­ replacement cornea was taken from another sideration, particularly as experiments and animal of the same species, brought in analogy have shown the feasibility of the wounded, but not quite dead; adhesion took operation." Bigger pursued corneal trans- VISUAL OUTCOME OF CORNEAL TRANSPLANTATION 267 plantation because of its potential as a treat­ vival of the patient correlates with graft sur­ ment for blindness and disability and he was vival. These grafts are done to overcome largely responsible for initiating the line of diseases which are inevitably fatal without researchers who have maintained the assault successful transplantation. on corneal transplantation for the last 150 Corneal transplantation is not done to pre­ years. serve life but to overcome disability, or to pre­ Great advances have been made since Big­ serve the integrity of the eyeball. Sometimes ger made his seminal contributions to corneal the procedure is carried out to reduce pain. transplantation. Anaesthesia, asepsis, and Ther� is, therefore, the need to invoke a effectiveantimicrobial and anti-inflammatory range of outcome measurements to correlate therapy have been introduced into surgical with the indications. Reduction of pain and practice. So too have microsurgical tech­ preservation of the globe are relatively easy to niques and materials, and the nature and sig­ judge and count.An assessment of the impact nificance of the allograft response have been of corneal transplantation on visual disability elucidated. Despite these giant strides in is a more complex consideration. medical science, not all of the problems of When considering the outcome of the pro­ corneal transplantation have been solved. cedure done for reduction of visual disability, Although the procedure has the potential to it is appropriate to consider a number of restore vision to those who are blind or objective measurements. Corneal graft func­ visually disabled as a result of corneal disease, tion is one important consideration. A cor­ there is a very wide spectrum of corneal dis­ neal graft must be thin and transparent if the orders and some patients gain more from patient is to see through the cornea. This is transplantation than others. Bigger recog­ not particularly useful information in itself. A nised that the major question was how to failed graft will always_ result in poor vision decide who would be helped by corneal trans­ but a functioning graft may not necessarily plantation, and 150 years on this still remains deliver good vision.A measure of visual func­ a major question. tion is to be preferred, but which clinical measure is appropriate? Corneal transplantation: Choosing the Visual acuity measured with a Snellen chart appropriate outcome measurement is very widely used and has gained almost uni­ Choosing an appropriate and quantifiableend versal acceptance. Snellen acuity is not a use­ point is a critical consideration in all scientific ful or complete measure of visual function assessment. The choice of an end-point when used alone or without stipulation.Quite depends on an understanding of the aim of the apart from the well known limitations of exercise. In surgery, this can be complicated acuity testing, considerations such as the time because one procedure is often done for more after surgery that the vision is measured, and than one indication, and because the mea­ the optical correction used during the test, are surable aspects of a result may not correlate important in the context of corneal transplan­ well with what the patient considers to be an tation. The final refraction cannot be appropriate outcome. Deciding on an appro­ achieved until all the sutures have been priate outcome for corneal transplantation removed.This is often not until well into the involves some difficult considerations along second year after surgery. Furthermore, the these lines. form of optical correction which yields the Corneal transplantation is the most widely best visual acuity may not be acceptable to the practised form of clinical transplantation.It is patient, who may have to function on a day­ generally perceived as being the most success­ to-day basis with something quite inferior. ful. This is so if the same criteria are used to Visual acuity should be tested with the patient assess the outcome of corneal transplantation using his everyday or socially-acceptable cor­ as are applied to other forms of clinical allo­ rection. Even when these requirements are grafting, such as kidney, liver, bone marrow met, the Snellen acuity measurement pro­ and heart and lung transplantation. For these vides a far from complete picture of a patient's forms of essential organ transplantation, sur- visual performance. Many grafts are done for 268 D. J. COSTER patients whose vision in the contralateral eye precious resources to be allocated on the basis is good. Presumably, the purpose of the pro­ of benefits derived for money spent. We live cedure is to provide improved binocular func­ in the age of "Outcomology".4 It behoves us tion, better field, and perhaps improved to come up with a range of objective measure­ stereopsis. Any improvement in binocular ments and consumer-based criteria for eval­ function is not assessible using acuity uating the success of our procedures, measurements alone. The acuity measure­ including corneal transplantation. ment may be quite misleading. Anisometro­ The need to develop appropriate measures pia in one form or another is relatively of outcome for corneal transplantation has common after corneal transplantation.A high prompted us to establish two data bases, the level of acuity may be achieved only with an Australian Corneal Graft Register, and the anisometropic correction. To assess the true Flinders Medical Centre data base, in order to meaning of an acuity reading requires a evaluate relationships between the various knowledge of the vision in the other eye and factors which influence outcome. the refraction of both eyes. Another approach is to disregard the Australian Corneal Graft Register results of psychophysical tests such as acuity The Australian Corneal Graft Register was and refraction, and to assess visual disability established in 1985. Ophthalmologists from directly. This approach takes surgeons out of aJl states of Australia contribute prospective the area of the objective measurements with data about their patients.) They provide data which they are familiar, into the more sub­ on standard forms at the time of surgery and jective, consumer-directed, approach of the foJlow-up data at regular prescribed intervals. market surveyor or sociologist. The process is More than 2,200 grafts are now being fol­ time-consuming, the results can be difficult to lowed. The data is analysed using customised interpret, and, perhaps more importantly, the software and a report is published each year. results must be interpreted individually for The Register provides very useful infor­ each patient because the expectations of mation because of its size and the breadth of patients having surgery for disability vary practice it describes. However, the deficien­ greatly. Social, economic, psychological and cies of such a data base must be acknowl­ cultural factors interact with physiological edged. About 80% of the grafts done in factors to determine how a patient expects to Australia are included, so the material con­ benefit from surgery.If we are to accept the tributed is not entirely comprehensive. There patients' judgement of whether our surgery may be differences in the way observers has reduced disability, we must make a careful report observations despite everyone's best assessment of their perceptions of their pre­ attempts to standardise observations and operative disability and post-operative expec­ recording procedures. Of course, facilities tations. Matching the patients' expectations and conditions vary from centre to centre and and the surgeons' expectations is a well-recog­ this may be particularly important when mak­ nised pillar of the act of medicine, but it is ing psychophysical measurements such as notoriously difficult to achieve with some visual acuity. patients. In order to take advantage of the strengths When deciding upon the criteria to be used of the Register and to minimise the weak­ to assess the outcome of corneal transplanta­ nesses, contributing surgeons can have their tion, it is not enough to consider the patient own results analysed separately and confiden­ and the surgeon. There are interested third­ tiaJly. They can also include information parties. Increasingly, medicine is coming related to their particular interests in addition under the watchful eye and guiding hand of to the core material which is standard for aJl those who finance health services. Govern­ contributors. This source of information is ment agencies and insurance companies are very important when dealing with highly increasingly interested in the outcome of specific outcomes such as visual acuity, where medicine in relation to resources consumed. standard observations are so critical. To look More and more, we can expect increasingly at more specific aspects of corneal transplan- VISUAL OUTCOME OF CORNEAL TRANSPLANTATION 269 tation requires detailed analysis of cases. This grafts, even if the primary indication was not is difficult to do in the Register because of its visual. By being "all-inclusive" we can look at size, the far-flung geographic distribution of a more heterogeneous group and get a better contributors and the different approaches to idea of the crude factors which influence cor­ care in the different institutions. To overcome neal transplantation in the general sense. this, some centres keep their own data, con­ centrating on aspects of care and outcome Visual acuity that they are interested in, or think they might The results of surgery, expressed in terms of become interested in, in the future. The visual acuity, are set out in Table II and Fig. 1. Australian Corneal Graft Register is happy to In view of the heterogeneity of the group, the manage this data on behalf of individual con­ results are surprisingly good. Approximately tributors. At Flinders Medical Centre, more 50% of grafted eyes achieved an acuity of 6/18 than 300 corneal grafts have been carried out or better with appropriate optical correction by the author and followed for nearly ten (not pinhole). years in some cases. Using data from this Of course this means that 50% of patients source, a number of questions can be ad­ do not achieve this level of vision. Approxi­ dressed that are difficult to answer with data mately 20% of the latter patients had a failed from the Register. graft and therefore had no chance of achiev­ ing an excellent visual result. Of the remain­ Indications for corneal transplantation der, 40% had co-existing ocular disease, such As previously mentioned, any assessment of as retinal disease or glaucoma, and 40% had outcome must bear a direct relationship to the an optical problem, such as astigmatism, to indication for surgery, and the indications for account for their poor vision. The corneal transplantation vary markedly from co-morbidities and complications recorded as one setting to another. The indications for contributory to a poor result are set out in corneal transplantation as recorded in the Table III. Australian Corneal Graft Register are set out in Table I. It is clear that the major indication Visual disability for surgery is to reduce visual disability, with Although Snellen acuity is the most widely the control of focal inflammatory processes or accepted measure of visual performance, and pain less frequent indications. notwithstanding the difficulties discussed Improvement in vision is the usual indica­ above, the relationship between visual acuity tion for corneal transplantation and will now and visual disability is not straightforward. be considered more thoroughly. There will be Visual acuity measurement has all of the prob­ no further discussion of the factors which are lems associated with psychophysical tests. associated with the outcome of grafts done to The measurement of disability is even more control focal destructive disease or pain. For complicated. It is not just a person's disability the purpose of this discussion, it is appro­ which affects his day-to-day functioning, it is priate to look at all patients having corneal his perception of his disability which is criti­ cal. If the patient cannot do more after sur­ Table I. Australian Corneal Graft Register Report gery than he could before, then the effort has 1989. Indications for graft in the Australian Corneal been wasted. With this bald judgement in Graft Register mind, the limitations of visual acuity as a Indication Per cent measure of the outcome of corneal transplan­ tation become even more obvious. The life­ Keratoconus 33% style of the patient, the expectations of the 23% Bullous keratopathy patient which are likely to be affected by the Failed previous graft 13% Corneal scars 12% pre-operative vision, and the attitude of the Corneal dystrophies 6% patient to life in general and vision in par­ Corneal ulcers 3% ticular, are all factors which will affect the per­ HSVand HZO 3% ceived outcome. An analysis of the results of Miscellaneous 7% corneal transplantation in terms of the visual 270 D.l. COSTER

Table II. Australian Corneal Graft Register Report 1989. The results of corneal transplantation expressed in terms of visual acuity

Number of patients Surviving grafts Failed grafts Best corrected visual acuity -PH' +PH -PH +PH

6/4 9 6/5 38 6/6 171 12 6/9 155 19 1** 6/12 75 34 6/15 1 6/18 63 32 4 6/24 32 16 6/30 2 6/36 31 14 3 1 6/60 30 20 6 6 61120 3 1 1 Count fingers at 3 feet 35 13 30 7 Hand movements 14 4 31 Light perception only 1 2 7 No light perception 2 10 Unknownlnot recorded 97 30

TOTAL 756 169 123 16

'-PH: minus pinhole; +PH: plus pinhole; **: IOL touch to cornea

acuity achieved in the eye after surgery, neg­ With the aim of uncovering some of the fac­ lects these considerations. tors affecting the visual outcome of corneal This is not to say that the achievement of a transplantation in the broader sense, we have clear functioning graft with good acuity is not looked at a group of patients under our own a high priority.It does, however, suggest that care at Flinders Medical Centre. A group of achieving a clear graft, or even a good visual patients who had received grafts were recalled acuity, is not the whole story. To have any and the status of their grafts, visual acuity and chance of improving the patients' lot, a graft refraction, visual disability and attitudes to must function, and with minimal optical aber­ the surgery, were evaluated. This evaluation rations, and the eye with the graft must be free was carried out on patients who had been of any other disorder which could limit visual operated on more than two years previously, potential. Furthermore, the patients' lifestyle in the expectation that their refraction and must be capable of benefiting from the rehabilitation would be complete. The study improved vision. will be reported elsewhere,6 but some of the more important features are relevant to this

I2Z2l SuNOvina groft (Wp ..... ln_) discussion. � SUrvMnggraft (with pinhOle) � Failed graft (WIth. wtthOUlp'nhole A consecutive series of 209 cases fulfilled the criteria with respect to time. However, Table III. Australian Corneal Graft Register Report 1989. Ocular decrease contributing to visual acuities less than 6118 in patients with clear grafts

Factor affecting visual outcome Number (%)

Cataract 36 (3%) BEST-CORRECTED VISUAL ACUITY (GRAFTEDeVe) Amblyopia 24 (2%) Aphakia 50 (5%) Australian Corneal Graft Register The Fig. 1. 1989. Retinal detachment 12 (1%) results of corneal transplantation expressed in terms of Cystoid macular oedemalmaculopathy 120 (11%) visual acuity. VISUAL OUTCOME OF CORNEAL TRANSPLANTATION 271 patients who lived more than 500 kilometres used contact lenses on the grafted eye, with a away were excused, as were those who were further 20% using both spectacles and contact infirm, unable to travel to the clinic, or who lenses. Only 8 (13%) had required keratore­ had died in the interim. This left a group of 60 fractive surgery to reduce spherocylindrical patients to study. They were evaluated by a astigmatism. person from outside the clinical team, a The questionnaire was directed towards medical student on elective secondment. evaluating the level of disability patients had Patients were re-assessed with a clinical exam­ following surgery and their attitude to their ination, including Snellen acuity, reading-line corneal transplantation. Approximately 75% test and keratometry, and were given a stan­ of graft recipients reported that they were dard questionnaire aimed at discovering their managing their day-to-day activities well. level of visual disability and their attitude to However, some patients commented that the corneal transplantation procedure. lighting conditions affected their ability to Because the study was a preliminary to more function and that glare was a particular prob­ extensive outcome studies, no attempt was lem. Forty per cent of patients felt restricted made to limit the indications for surgery. with respect to hobbies and pastimes. About half the patients (48%) had surgery for Although the overwhelming majority (93%) keratoconus. Nine patients (15%) were oper­ believed that the graft had been "worth the ated on for a previously failed graft. The trouble" and they would be prepared to go remainder had the sequelae of inflammatory through the procedure again, less than half disease and two cases were operated on for believed that the outcome had matched their impending perforations. expectations. Approximately 65% achieved 6/18 vision or There was a significantcorrelation between better in the grafted eye (Fig. 2). In 52% of a patient's level of satisfaction and whether patients, the level of vision achieved was bet­ the visual acuity in the grafted eye matched or ter than in the contralateral eye. In 13% of exceeded the contralateral eye. Furthermore, patients, the grafts had failed. Of those that satisfied patients tended to report that their were clear and functioning, 22% had more preoperative expectations were met. than 5 diopters of spherocylindrical astigma­ It is usually accepted that improvement in tism. A further 13% had irregular astigma­ medicine can come from a close evaluation of tism (Fig. 3). Spectacles were used by 52% of failures and the identification of the reason patients to achieve their best vision and 17% behind the failure. It was considered that SNELLEN ACUITY - GRAFT K READINGS IN GRAFT

20 .------, 12 �------,

10

rn 8 I- � Z Z UJ w 10 6 i= �

o o ..-NC').., ."fD,.. CDc»O..- NN:::E

O�c:.c:,.';'cD,!,cD�":""':�-0>0..­ ..- K READINGS SNELLEN ACUITY

Fig. 2. Flinders Medical Centre Data. The results of Fig. 3. Flinders Medical Centre Data. The results of corneal transplantation in sixty patients expressed in corneal transplantation in sixty patients expressed in terms of visual acuity. terms of keratometry readings. 272 D.J. COSTER eight patients indicated that they were signifi­ increased dramatically in recent times, and, cantly dissatisfied with the outcome of the along WIth. this, the management of the allo­ procedure. Three of these patients had failed graft reaction and related problems at a clini­ grafts. Of the other five, the Snellen acuities cal level is very much improved. with correction were remarkably good. Four The main strategy for decreasing the impact of them had vision of 6/12 or better and one of the corneal allograft response is to avoid had vision of count fingers which was attribut­ doing grafts in high-risk cases unless there are able to astigmatism and an inability to cope real improvements in lifestyle to be gained. with any form of correction. All of these five When corneal transplantation must be carried patients reported that they disliked wearing out on such cases, one must be prepared to do contact lenses and indeed this factor appeared all that is necessary to reduce the incidence of to be the major source of their unhappiness. rejection. Grafts done for inflammationor the seque­ Strategies for improving the visual outcome lae of allograft rejection are prone to rejec­ of corneal transplantation tion. Quite often this inflammatory disease For a patient to benefit from corneal trans­ only affects one eye. In such cases, vision in plantation, the graft must function in the the other eye is excellent. Corneal transplan­ metabolic sense; it must be o�tica\ly clear tation should be avoided in such cases when­ with minimal aberrations, the eye should be ever possible. free of co-existing disease which might reduce When it is necessary to offer corneal trans­ the visual potential below what would other­ plantation to patients with inflammatory wise be expected, and there needs to be room disease, there are a number of things that can for improvement in the patient's lifestyle. be done to reduce the chance of graft failure. Therefore, the outcome of corneal transplan­ The single most important step in preventing tation can be improved by reducing the allograft rejection is to minimise post-oper­ impact of corneal graft failure, reducing the ative inflammation. After surgery there is impact of post-operative aberrations and co­ movement of cells from the limbus across the existing ocular disease, and ensuring that the host cornea to the host-graft junction. This procedure is restricted to those who, because influx 'of cells is necessary for wound healing of their disability, are likely to benefit from but also carries with it a cellular arsenal with the surgery. the potential to initiate the allograft response. \0 What is required, is enough cell Measures to reduce the impact of allograft migration to achieve wound healing but not rejection enough to initiate rejection. Pre-operative Corneal allograft rejection is the most com­ pathology and post-operative factors such as mon cause of corneal graft failure. The factors episodes of inflammation, from whatever influencing the survival of corneal grafts have cause, determine the tendency for cells to received considerable attention in recent accumulate at the graft-host junction. Grafts years. At a clinical level, vascularisation of done for keratoconus seldom become the recipient bed and inflammation, or a inflamed, are often very slow to heal, with history of inflammation, are well established instability at the graft-host junction for a year factors affecting corneal graft survival.7 At a or more after surgery, and seldom reject. histological level, survival is influencedby the Grafts done for inflammatory conditions such number of interstitial dendritic cells in the as herpetic keratitis heal very quickly and the recipient's corneal bed.s As with most other sutures can often be removed after only a few forms of allograft rejection, the degree of months; recurrent inflammation with sub­ major histocompatability antigen mismatch sequent allograft rejection is a common cause between donor and recipient influences graft of graft failure. The tendency to accumulate survival.9 Although a discussion of the cytoki­ cells can be modified to some extent by anti­ netics of the corneal allograft response is inflammatory therapy. Topical corticosteroid beyond the scope of this discussion, it is clear preparations are most potent in this regard. that our understanding of the process has Inflammation must be avoided by carrying VISUAL OUTCOME OF CORNEAL TRANSPLANTATION 273 out surgery in the most atraumatic manner, by Prevention of post-transplantation astig­ optimal use of topical corticosteroids, and by matism is a high priority for corneal trans­ promptly recognising and treating any post­ plantation surgeons. Refinements in cutting operative inflammation as might occur with techniques and suturing strategies have been acute blepharitis, conjunctivitis, uveitis or any developed with this in mind. The expectation other inflammatory condition, or as may has been that if a perfectly circular graft is occur should a suture become loose and placed in a perfectly circular defect, the sides exteriorised. of the graft and recipient defect have com­ In high-risk cases, those in which the blind­ plimentary configurations, and the suturing ing corneal disease has an inflammatory com­ pattern is perfectly regular with regard to ponent, it is much more difficult to avoid geometry and tension, astigmatism will be post-graft inflammation and subsequent allo­ eliminated. Either this level of precision has graft rejection. In addition to the measures not been achieved or the hypothesis is faulty. suggested above, histocompatibility match­ The surgical techniques proposed to reduce ing9 and systemic immunosuppressionll can astigmatism are numerous. In particular, vari­ reduce the incidence of rejection. Neither of ous patterns of suturing have been advocated. these optioos is simple or inexpensive. One approach that has received considerable Matching is difficultbecause of the extreme attention recently, and which has been quite polymorphism of the major histocompatibil­ widely adopted, has been the use of a combi­ ity complex, and the expensive facilities nation of continuous and interrupted sutures required to pursue this strategy.Nevertheless, with selected removal of the interrupted ele­ the benefits of matching for class I antigens, ments in order to reduce early post-operative though incomplete, are quite definite. astigmatism and bring about early post-oper­ ative rehabilitation.13 As Binder points out, Immunosuppression is not only expensive, it is potentially dangerous. It can only be con­ the effect of selective removal of sutures is lost sidered for patients who are blind because after all sutures are removed and the advan­ tage is apparent only in the firstyear after sur­ they need a corneal graft and who have been gery. This advantage in the first year after well informed about the complications of surgery must be balanced against the disad­ systemic immunosuppression. The use of vantage of interrupted sutures. All suture systemic cyclosporin has greatly improved the materials, including monofilament nylon, are effectivenessof pharmacological immunosup­ inflammatory to some degree and for some pression for essential organ transplantation.1 2 reason new vessels are encouraged by the Topical preparations have yet to be shown to presence of knots. Furthermore, interrupted be effective in a clinical situation. 'i>utme'i> can be more ditflcu\t to remove than a Measures to reduce the impact of refractive continuous nylon suture and small fragments aberrations of nylon may be left behind in the cornea.The A clear graft will not result in good vision if occasional tendency for vascularisation the optical properties of the cornea are towards the graft margin may be a high price inappropriate. Astigmatism is a common to pay for the temporary reduction of astig­ observation and may be regular (sphero­ matism in the early post-operative period. cylindrical) or irregular in form.Anisometro­ Since all surgeons continue to get some pia due to the grafted cornea having astigmatism in their grafts, although perhaps significantlydifferent optical properties to the less today than in the past, it remains impor­ cornea in the contralateral eye is also tant to lessen the impact of astigmatism on the common. visual outcome of transplantation. Careful selection of appropriate optical aids is a very Prevention of surgically-induced astigmatism important element in the management of all Post-operative optical correction can be very patients receiving corneal grafts. Most cases difficult, and prevention of post-corneal of post-transplantation astigmatism can be transplantation refractive errors is the pre­ managed with spectacles or contact lenses. ferred option. However, it is also difficult to However, deciding on the appropriate correc­ achieve. tion demands considerable skill. 274 D. J. COSTER

Refraction after corneal transplantation fully appreciated with photokeratoscopy or Achieving the full visual potential of a corneal with the use of other more complex systems graft may be as large a problem as achieving for demonstrating the three dimensional and maintaining a clear graft. Of course the characteristics of the cornea. A number of preferred outcome is that a graft is so physio­ highly sophisticated systems have been devel­ logical that no optical correction is required to oped to evaluate the complex relationship achieve a high level of binocular vision.When between variations in corneal shapes and opti­ this has not been achieved, spectacles are the cal performance. These systems are among next most acceptable solution, but sometimes the tangible fruits of the development of cor­ it is necessary to use contact lenses to achieve neo-refractive surgery. They work on a the full visual potential of the operated eye. number of different principles, provide infor­ It is important for the patient to understand mation in various forms, and are invariably the complex nature of the process of refrac­ expensive. It is not clear that they provide tion and rehabilitation, and in particular that information about the topography of corneal the process must be repeated a number of grafts that cannot be achieved by obscuril}g times in the post-operative period and that a the pattern of fluoresceinunder a rigid contact stable result will usually not be achieved lens. A flowchart for refracting patients with within 18 months of surgery. The prolonged corneal grafts is set out in Figure 4. nature of the process and the unphysiological Contact lens fitting shape of the corneal graft contribute to the When the topography of the grafted cornea is overall difficulty of achieving a satisfactory irregular, it is necessary to use a contact lens optical correction for patients who have had to achieve the full visual potential. In selec­ corneal grafts. ting the appropriate lens design, the central Although the principles of refraction apply corneal curvature as determined by keratom­ as much to patients with corneal transplants as etry, the topography of the optical zone of the anyone else, some additional information is central cornea, the wound edge configura­ often helpful. It is important to utilise infor­ tion, and the relationship of the graft curva­ mation from the distance and near acuity, pin­ ture to the host curvature are the critical hole acuity, retinoscopy, keratometry, and factors.Irregular astigmatism can only be cor­ photo keratoscopy. Irregular astigmatism is rected with a relatively rigid contact lens. common after corneal transplantation and is However, a more flexible, more forgiving often the reason why it can be so difficult to material may be required to overcome irregu­ correct a patient's acuity up to the level larities in topography at the graft-host junc­ achieved with a pinhole.The nature of irregu­ tion.In such circumstances, a soft lens may be lar astigmatism can be appreciated with ret­ employed but if there is also central corneal inoscopy and keratometry, but can only be irregularity a soft lens will not deliver good DtlCteased Visual Acuity vision. Under these circumstances it may be necessary to employ a system of soft lens and

. 9.5·10.0mm OS L;-; iApprox I- DlAGN �n � BOZR-FjattestK l -=r ,- - I __ Abnorma lGraft __ � __ NormalGraft I Edge Contour I Edge Contour Regular Central Cornea Irregular Central Cornea Largel0mrn L_ �- I GP LENS BOZ� FlattestK iGPLEN I GPLENS 6 BOZO Graftd(a. +1.0mm L_----; ! __�-" 1 I Large 10 mrn BOZR Flattest K Uncomfortable UncomfortablCt BOZO Graftdia.+l.0mm Tri-quadracurve , __1- SOFT CLt\-=-l "PIGGY BAC�" I SOFTPERM

Fig. 4. An algorithm for refraction of patients with Fig. 5. An algorithm for the fitting of contact lenses to corneal grafts. patients with corneal grafts. VISUAL OUTCOME OF CORNEAL TRANSPLANTATION 275 an over-riding hard lens centrally. Recently, For higher levels of astigmatism, wedge hybrid lenses have been introduced which resections have been employed. These pro­ have a progression of rigidity from the periph­ duce a greater degree of correction but the ery to the centre. Softperm is one such lens results are less predictable. Again, using a which has a definite but limited place in the standard procedure of two wedges in the flat­ rehabilitation of patients who have had cor­ test meridian, each with a width of 1 mm and a neal grafts. Unfortunately, the current lenses length of three clock hours, we achieve an of this type are not high performance lenses in average correction of nine diopters when all the physiological sense and this limits their sutures are removed.It is necessary to achieve usefulness in the context of corneal transplan­ over-correct initially because invariably some tation. A flow chart setting out the principles correction is lost when the sutures are of lens fitting and the usual lens parameters removed. for patients with corneal grafts is set out in Figure 5. Anisometropia after corneal transplantation Astigmatism is not the only troublesome Surgical correction of astigmatism refractive problem for patients having corneal Corneal astigmatism is the usual but not the grafts. Anisometropia can be particularly only indication for contact lens fitting after troublesome if the principal aim of the corneal corneal transplantation. Anisometropia can transplantation is visual and the patient is also be a significant post-operative problem, aphakic, pseudophakic or needs cataract sur­ particularly when the surgery is done to gery along with the corneal graft. In many improve binocular function. communities, pseudophakic bullous kera­ Correction with spectacles and contact topathy is the commonest indication for cor- lenses is not always possible despite the best neal transplantation. Under these efforts of the most experienced practitioners. circumstances, avoiding anisometropIa Not only does the post-operative optical cor­ becomes an important consideration. When rection need to deliver excellent vision, it the crystalline lens has been removed or needs must be acceptable to the patient. Whether to be removed in a patient with a corneal the form of correction is acceptable is deter­ graft, the options for avoiding anisometropia mined to some extent by the patient's physi­ are restricted to selecting an appropriate ology but more by the patient's lifestyle and power for the intraocular lens. Very little can attitudes. When it is necessary to improve be done to control selectively the spherical visual performance beyond what can be refractive power of the corneal transplant. achieved with optical devices acceptable to Since aphakic contact'lenses, even of the most the patient, and there is residual regular cor­ sophisticated kind, impose a considerable neal astigmatism, surgical options are avail­ physiological stress on the corneal graft, an able. Various approaches have been intraocular lens is the preferred option. A advocated over the years including relaxing posterior chamber lens is desirable since the incisions, wedge resections, non-radial ker­ survival of corneal grafts is much better with a atotomy and regrafting.14 posterior chamber lens than with an iris fix­ Relaxing incisions placed in the axis of the ated or anterior chamber lens (Fig. 6). greatest curvature have proved to be the most This has some implications when grafts are effective. This simple procedure is safe and done for pseudophakic bullous keratopathy. predictable. Of the corneal grafts done at Whenever possible, posterior chamber lenses Flinders Medical Centre, 8% go on to have should be left in situ, and iris clip lenses or relaxing incisions. A standard procedure is anterior chamber lenses are best removed if used with two deep incisions, as deep as one this can be achieved without significant can safely go, bearing in mind the unpredict­ trauma. Whether posterior chamber lenses able thickness of the graft host margin, each sutured into the ciliary sulcus have a compar­ incision being three clock hours in length.The able prognosis with regard to graft survival as average correction with this procedure is those supported by an intact lens capsule, slightly less than five diopters. remains to be determined. Selection of the 276 0.1. COSTER

-' coma and retinal conditions, can also account � 1.0 for a poor visual result despite a functioning �_____ PC IOL (n = 169) � en 0.8 graft. The conditions which have contributed to poor visual results in the group of patients

and insertion of an intraocular lens, an argu­ -' < ment can be made for staging the procedure > 1.0 :; Triple procedure (n=98) so that the curvature of the corneal graft can II: � be used to predict accurately the lens power en 0.8 ... U. IOL Inserted aller grail (n=72) required to avoid anisometropia. The risk of < II: 0.6 this approach is in jeopardising the graft's sur­ (!) U. vival with the subsequent surgical procedure. 0 However, it appears that a subsequent catar­ > 0.4 � act extraction and lens implantation does not iii < 0.2 significantly alter graft survival (Fig. 7). m 0 II: Whether this approach provides a better c.. visual outcome with regard to binocularity has o 1 234 not been determined. TRIAL TIMES (YEARS POST-GRAFT) Fig. 7. Australian Corneal Graft Register Report Measures to reduce the impact of co­ 1989. Survival of corneal grafts in patients who have morbidities and complications on the had triple procedures or lens insertion as a secondary outcome of corneal transplantation procedure. There is no statistical difference between the Co-existing disease, such as cataract, glau- two groups. VISUAL OUTCOME OF CORNEAL TRANSPLANTATION 277 value of the procedure. Patients for whom the patients who need contact lenses to achieve post-operative result did not meet their pre­ the desired level of vision after a graft but operative expectations, tend to be unhappy. who, for one reason or another, are unable to So too are patients who have negative atti­ tolerate lenses are also unhappy graft reci­ tudes to contact lenses and end up requiring pients. Since as many as 15% of graft patients the devices to achieve their best vision. will need contact lenses after surgery to That a person's attitude affects the way he achieve optimal vision, patients with an aver­ copes with illness, and that depression may sion to contact lenses should be avoided have a devastating effect on the way a person wherever possible. is affected by illness, is well known to all clin­ Some patients present for corneal trans­ icians. It is also generally well appreciated plantation because they are averse to wearing that improving a patient's attitude to illness, contact lenses. These patients are likely to be and therefore his ability to cope, can be very unhappy if they need to wear lenses post­ difficult. For patients having corneal trans­ operatively to correct astigmatism or aniso­ plantation, the problem is a little easier to metropia. Often a re-education programme address because the attitudes which require prior to surgery, so that lens wear is tolerated, attention are more readily identified. can be a worthwhile investment and avoid Considerable effort must be made to ensure much unhappiness after surgery. that the patient's expectations and the sur­ geon's expectations match. To achieve this The importance of case selection takes a long time and is usually beyond the Corneal transplantation is a remarkably suc­ personal resources of most surgeons. The task cessful procedure, but, as with most other can be simplifiedby using trained counsellors, surgical procedures, there is room for distributing reading materials, using audio improvement. Patients expect that their sur­ tapes and video material. Whatever approach geons will deliver meticulous surgery which is employed, it is important for the surgeon will minimise the risk of complications and and patient to come together prior to surgery maximise the visual results of surgery. These to ensure that the surgeon's expectations and are understandable and reasonable expec­ the patient's expectations are the same. Fail­ tations, but outcomes are not invariably suc­ ure to achieve matched expectations can often cessful. There are some unhappy patients. result in a patient being dissatisfied and Over the years, a powerful approach to unhappy. Even when surgery is perfect and improving the effectiveness of medicine and the case selection, based on medical grounds, surgery has been a close examination of our is appropriate and when the outcome has mistakes and unsuccessful undertakings. If we matched the expectations of surgery, there are are prepared to assess surgical procedures still disappointed patients for whom the done for disability, including visual disability, results have not fulfilled their own expec­ in terms of the impact on our patients' lives, tations. Disparities between the expectations and then to allow our patients to make this of surgeons and patients are a major cause of judgement, we are shown the way forward. unacceptable results for patients. Every effort The most powerful approach to improving must be made to inform the patient of the pro­ the outcome for our patients is to be even cess of surgery and after-care and to describe more stringent in our selection criteria. to them the likely outcome. With all the will in Patients who achieve better vision in the oper­ the world, this cannot always be achieved as it ated eye than the contralateral eye are almost is the inherent attitude of patients which influ­ invariably pleased with the results of surgery. ences the results. The assessment of patients' Surgeons who wish to work exclusively with attitudes and of ways to direct them benefi­ satisfiedpatients and who wish to ensure that cially is the corner stone of the art of medicine precious resources, including their own time, and is not readily amenable to quantification are used to maximum advantage should avoid and experiment. operating on patients in whom a corneal graft Attention must also be directed to the is unlikely to improve vision beyond what is patient's attitude to contact lenses. Similarly, present in the other eye. 278 D. J. COSTER

If there is one lesson that has to be learned SM, Mahmood M, Coster DJ (on behalf of all and relearned, it is the need to select patients contributors): The Australian Corneal Graft Reg­ istry, 1989 report. Adelaide . Flinders Press, 1990. carefully. This was well recognised at the 6 Williams KA, Ash JK, Pararajasegaram P, Mills R, dawn of corneal transplantation, before the Harris S, Coster DJ: Long-term outcome after procedure had been introduced into clinical corneal transplantation: visual result and patient practice. In his paper of 1837, Bigger wrote perception of success. Ophthalmology (In Press). 7 Volker-Dieben HJ, D'Amaro J, Kok-van Alphen " ...the operation should not be sanctioned CC: Hierarchy of prognostic factors for corneal under any circumstances, when the patient allograft survival. Aust NZ J Ophthalmol 1987, enjoys even a tolerable degree of vision with 15: 11-18. the other eye, at least until our knowledge has 8 Williams KA , White MA, Ash JK, Coster DJ: Leu­ kocytes in the graft bed associated with corneal been increased by further experiments and graft failure . Ophthalmology 1989, 96: 38-44. observations." These words remain true to 9 Sanfilippo F, MacQueen JM, Vaughn WK, Foulks this day. GN: Reduced graft rejection with good HLA-A and B matching in high-risk corneal transplanta­ tion. New Engl J Med 1986, 315: 29-35. Key words: corneal transplantation, visual outcome, 10 Williams KA, and Coster DJ: The role of the limbus refraction, contact lenses, rehabilitation. in corneal allograft rejection. Eye 1989, 3: 158--66. 11 Hill JC: The use of cyclosporine in high-risk ker­ References 1 atoplasty. Am J Ophthalmol 1989, 107: 506-10. Blainey G: A game of our own-the ongms of 12 Borel JF: The cyclosporins. Transplant Proc 1989, Australian Football. Melbourne. Information 21: 810--15. Australia, 1990. 13 2 Binder PS: Selective suture removal can reduce Darwin E: Zoonomia or the laws of organic life . Vol postkeratoplasty astigmatism. Ophthalmology II. London. J. Johnson, 1796. 1985,92: 1412-16. 14 3 Bigger SL: An inquiry into the possibility of trans­ Lavery GW, Lindstrom RL, Hofer LA , Doughman planting the cornea, with the view of relieving DJ: The surgical management of astigmatism blindness caused by several diseases of that struc­ after penetrating keratoplasty. Ophthalmic Sur­ ture. Dublin J Med Sci II: 408-17. gery 1985, 16: 165-9. 4 15 Zimmerman TJ : Outcomology. Editorial. Arch Meyer RF, and Musch DC: Assessment of success Ophthalmol 1990, 108: 342-3. and complications of triple procedure surgery. 5 Williams KA, Sawyer MA, White MA, Muehlberg Am J Ophthalmol 1987, 104: 233-40.