THE ROLE OF LIGHT SCATTER IN THE DEGRADATION OF VISUAL PERFORMANCE BEFORE AND AFTER Nd:YAG

R. R. GOBLEl, D. P. S. O'BRARTl, C. P. LOHMANNl, F. FITZKE2 and J. MARSHALLl London

SUMMARY high but some complain of glare around point sources of

The aim of this study was to determine whether capsulo­ light at night, and a few consider it sufficiently disturbing tomy size influences visual performance. Snellen visual to cause them to refrain from driving at night. No syste­ acuity and forward light scatter (light scattered towards matic study has been undertaken to evaluate the effects the patient's , but out of the focussed retinal image) that surgical parameters such as capsulotomy size and measurements using simple computer graphics based on implant chipping may have on the patient's post-operative van den Berg's technique were used to measure visual visual performance. performance. Twelve patients were studied: 4 had small [n many clinical practices. the only assessment of visual central through undilated pupils and 8 performance is Snellen acuity testing. Few centres per­ had wide capsulotomies through dilated pupils. The two form variable contrast acuity measurements although sev­ groups were matched for age and pre-laser Snellen eral such tests are now readily available.' In practice if a acuity. Following treatment, both groups had equal patient complains of visual deterioration or the corrected improvements in Snellen acuity. There was only a signifi­ Snellen acuity is noted to have worsened. then ocular cant (p

Eye (1994) 8, 530-534 © 1994 Royal College of Ophthalmologists INFLUENCE OF CAPSULOTOMY SIZE ON VISUAL PERFORMANCE 531 pseudophakic patients. Individuals were assessed both geometry was employed, the vertical exposure was before and after Nd:Y AG laser capsulotomy. increased to 5-6 mm and the horizontal exposures to the capsulotomy edge often needed to be multiple. If capsule MATERIALS AND METHODS flaps remained adherent to the implant, a small circumfer­ ential exposure at the capsulotomy edge usually sufficed Patients to remove them from the central area. The average number Patients were selected from those referred for Nd:YAG of pulses was 88, the average pulse energy was 1.0 ml and capsulotomy. In order to minimise fluctuations in results the average total energy was 94 ml. resulting from collateral ocular pathology, the following exclusion criteria were employed: media opacities other Measurement of'Snellen Acuity than opacified posterior capsules (e.g. anterior capsule Following retinoscopy and appropriate refractive correc­ remnants), pre-existing macular disease (e.g. age-related tion with trial lenses, the patients viewed an internally ill­ macular degeneration) or optic nerve disease (e.g. glau­ uminated Snellen acuity wall chart at 6 m. The eye not coma), amblyopia or decentred intraocular lenses (e.g. being tested was occluded. All patients were tested under dialling holes or optic edges within or adjacent to the standard clinic conditions. pupil). This helped to eliminate other sources of forward light scatter/reduced visual performance. The actual test to Measurement of' Forward Light Scatter measure forward light scatter was designed for a popula­ The method employed in this study was a modification of tion of young myopes and a proportion of the elderly van den Berg's technique, using a computer graphics population in this study were unfortunately unable to per­ video display and computerised data collection and analy­ form or apparently to understand this test and had to be sis.7.x The retinal light distribution from any visual excluded.6 environment consists of two components: the focussed Twelve (11 patients) aged between 60 and 82 years image of the object of interest and unfocussed light from were assigned to one of two groups. Group A consisted of the environment, the latter producing a more homog­ 4 eyes of 4 patients who underwent small central capsulo­ eneous background level upon which the focussed image tomies which were fashioned through undilated pupils is superimposed. If the background level increases relative (average age 74 years). Group B comprised 8 eyes of 7 to the image of interest, then contrast is lost and acuity patients in whom wide capsulotomies were performed deteriorates. Any component which scatters light out of after pupil dilatation (average age 71 years). All had been the focal point into the background serves to decrease the operated on at St Thomas' Hospital 6-34 months pre­ contrast between the two components and degrade vision. viously. The implants used in the two groups were all Van den Berg's test utilises this principle by surrounding a manufactured by Kabi Pharmacia and the details are given central target with an annulus of background illumination in Ta ble 1. and by comparing the contrast between the two a measure of intraocular scatter can be obtained. Capsulotom}' Te chnique Details of the computer technique have been described All capsulotomies were undertaken by R.R.G. or in full elsewhere.7 Stimuli were generated by a Viglen D.P.S.O'B. using a Lasag Microruptor MR-2G, operating IV /33 computer on a Viglen CA 1428B high-quality video in the Fundamental Mode with a 12 nanosecond pulse and graphics display. The computer screen was viewed mon­ fo cussed with the aid of a corneal contact lens. The tech­ ocularly at a fixed distance of 28 cm from eye to monitor. nique for laser capsulotomy was standardised for each The central fixation stimulus consisted of a circle with a group and is given below. radius of 1°. It was set at a level termed the match lumi­ In group A the pupil was not dilated and the exposures nance. This luminance was a random value brighter than were delivered initially over a vertical line typically that of the background. The central stimulus flickered at extending 2-3 mm in length and commencing inferiorly. 7.5 Hz between the match level and the background level. If capsule flaps remained within the pupil, the opening in The match level was under the control of the patient and the capsule was extended with repeated exposures in the could be increased or decreased by depressing two buttons horizontal meridian. The average number .of pulses was on the computer keyboard. The patient's task was to adjust 41, average pulse energy was 1.2 ml and the average total the contrast between the match luminance and the back­ energy required was 49 ml. ground in order to minimise or abolish the flicker. Once In group B, the patient's pupil was dilated with guttae this had been achieved a button was depressed and the tropicamide 1 % and, although the same exposure datum recorded. The computer recorded the match lumi­ nance value and the difference between the match and Table I. Kabi Phannacia intraocular lenses (lOL) w,ed in the study population background luminance values which were converted to contrast using the Michelson contrast definition.9 The IOL Group A Group B IOL characteristics match level was then automatically reset to a new random

720A 3 patients 4 patients 6.5 mm optic. all PMMA level and the process was repeated ten times. 725 Ale I patient 3 patients 7.0 mm oplie. all PMMA The test was divided into two parts: the first without UI37L None I patient 7.0 mm optic. 3-piece stray light and the second with a stray light annulus which 532 R. R. GOBLE ET AL. flickeredin counterphase to the match stimulus. The stray­ terior capsular opacification varies in the literature light contributed to the luminance of the central test target between 9% and 51 %, depending on the criteria used for due to forward light scatter and this increased the patient's diagnosis, the length of follow-up and the surgical tech­ error in attempting to abolish the flicker. The difference nique.IO.11 Some attempts have been made to relate between the two parts of the test provided a measure of the patients' visual disability to the morphological classifi­ level of forward-scattered light experienced by the patient. cation of capsule opacities. Fibrous plaques were claimed to produce simple attenuation of light entering the eye, Clinical Assessment after Capsulotomy whereas Elschnig's pearls and non-uniform clouding pro­ Measurements of best corrected Snellen acuity and for­ duced glare symptoms.12 The early studies relied on ward light scatter were undertaken immediately prior to measurement of glare disability. A simple method of and a fortnight after capsulotomy . On review after laser assessing visual deterioration in the face of a glare source treatment, patients were examined on the slit lamp without is to determine visual acuity with and without a pentorch mydriasis and the presence of anterior chamber flare or shone obliquely at the patient's eye.13 Other more accurate cells, capsule remnants within the pupil and evidence of devices have been developed, but rely on relatively intense intraocular lens 'pitting' by the laser exposures was noted. light sources to induce glare.7 The only methods for for­ ward scatter testing under photopic conditions rely on van RESULTS den Berg's direct compensation technique.� The develop­ Clinical Assessment ment of Lohmann et al. 's computerised system for assess­ ment of forward light scatter has allowed these tests to be No eye had clinically apparent uveitis on review after laser rapidly and repeatably carried out. treatment. Residual capsule flaps were noted in all 4 eyes Studies of variable contrast acuity in patients with tran­ from group A, some of which were fairly large. Two eyes sient corneal haze after excimer laser surgery demon­ from group B had small capsule tags remaining in the strated that only low contrast acuity was significantly pupillary zone. Two eyes from group A (n = 4) had central reduced post-operatively. This reduction in visual per­ pitting and 2 from group B (n = 8) had minimal peripheral formance showed no correlation with the measurements pitting of the implant by the laser. of reflected light, but almost absolute correlation with Snellen Acuity measurements of scattered light.6.7 Although measure­ ment of 100% contrast Snellen visual acuity is the stan­ The results of Snellen acuity measurement are shown in dard clinical assessment of central visual performance, it Fig. 1. Both groups had equal improvements in acuity fol­ is of limited value in predicting loss of visual performance lowing capsulotomy with 75% of patients displaying in patients with capsular opacification.14 It is for this improvements of at least one line. Three patients had no reason that all of our pre-laser patients demonstrated poor change in their acuities post-laser (I from group A and 2 results in forward light scatter assessment despite having from group B). There was no significant difference relatively good Snellen acuities. The results from this between the two treatment groups, as can be seen by the study showed that the size of caps ulotomy did not appear following average acuity values: to influence average Snellen acuity which improved from Group A pre-laser Group B pre-laser just less than 6/12 to better than 6/9 in both groups. In average: 6/13.04 average: 6/12.00 terms of the elderly patients that were being treated, this

Group A post-laser Group B post-laser improvement in acuity was felt to be clinically significant. average: 6/8.45 average: 6/8.22 The initial exposures with the laser were the most likely to produce lens pitting as the capsule could be closely Forward Light Scatter adherent to the posterior surface of the implant. Once the Clear differences were observed between the two groups capsulotomy had been started, the tension within the cap­ in measurements of forward light scatter. Group A showed sule was relieved and a gap often appeared between lens little improvement (Fig. 2). Group B displayed a marked and capsule. This reduced the risk of lens damage. By improvement which was statistically (unpaired Students initiating the capsulotomy at the inferior periphery of the t-test) significant at p

Patients GROUP A

2

• BEFORE LASER

Ii AFTER LASER

0 +---+ 6/24 6/18 6/12 6/9 6/6 6/5 Snellen Acuity

Patients GROUP B 4 3 • BEFORE LASER 2 m AFTER LASER

o 6/24 6/18 6/12 6/9 6/6 6/5 Snellen Acuity

Fig. 1. Snellen acuity in groups A and B he/ore and after Nd:YAG capsulotomy.

GROUP A % CONTRAST GROUP B 80 % CONTRAST )( 70 70 )( .. 50 )( 60 )( )( )( )( )( IE 50 )( I 50 )( )( • )( 40 )( 40 )( I )( II )( 30 30 )( )( • 20 20 )( II II 10 10 • � 0 K No Strayhght With Straylight No Strayllght With Straylight No Strayllght With Strayllght No Strayllght With StrayJight BEFORE LASER AFTER LASER BEFORE LASER AFTER LASER

Fig. 2. Forward light scatter measurements in patients of Fig. 3. Forward light scatter measurements in patients of group A hefore and after Nd:YAG capsulotomy. Each data point group B hefore and after Nd:YAG caps ulotomy. Each data point represents a single patient's eye. represents a single patient's eye.

received wide capsulotomies, there were dramatic that a capsulotomy should be at least as large as the sco­ improvements in percentage contrast error scores. In the topic pupil in order to allow it to regulate image intensity group with small central capsulotomies there was no sig­ and reduce diffraction. The factors that influence the size nificant improvement in forward scatter measurements of a capsulotomy include the tension within the capsule, and even the firstpart of the test (without stray light) failed the pulse energy in relation to the thickness of the capsule to improve significantly. This was due to the relatively and the placement of the laser bursts by the surgeon. If large amounts of residual capsule which remained within capsule remnants remain withinthe pupil margin they will the pupil to produce an irregular artificial entrance aper­ continue to produce light scatter and increase background ture and give rise to light scatter. The higher incidence of luminance with loss of the contrast of the focussed retinal central lens pitting in the latter group may also have con­ image. Pitting of the central optic zone by the laser will tributed in some cases. also produce forward light scatter and the risk may be There would appear to be a definite visual benefit in reduced by initiating the caps ulotomy at the lens creating a wide capsulotomy. Holladay et al.ls theorised periphery. 534 R. R. GOBLE ET AL.

Creating a larger capsulotomy may theoretically second Nd:YAG laser in 6664 cases. Am Intraocul Implant increase the risks of development of cystoid macular Soc 1 1984; 10:35-9. 5. Lohmann CP, Gartry DS, Kerr-Muir MG, Timberlake GT, oedema (CMO) and retinal detachment after Nd:YAG Fitzke F, Marshall 1. 'Haze' in photorefractive keratectomy: caps ulotomy. The barrier effect of the posterior capsule its origins and consequences. Lasers Ophthalmol1991;4: 15. prevents vitreous access to the anterior chamber and a 6. Lohmann CPo Gartry DS, Kerr-Muir MG. Timberlake GT, small capsulotomy may be a more effective barrier.16-1� Fitzke F, Marshall 1. Corneal haze after excimer laser refrac­ Clinical studies have not quoted the incidence of vitreous tive surgery: objective measurements and functional impli­ cations. Eur 1 Ophthalmol 1991;I: 173. in the anterior chamber or the size of caps ulotomies in 7. Lohmann CP, Fitzke F. O'Brart D, Kerr-Muir M, Timber­ those patients unfortunate enough to develop such compli­ lake G, Marshall 1. Corneal light scattering and visual per­ cations after Nd:YAG laser capsulotomy. The greater fonnance in myopic individuals with spectacles, contact number of laser pulses and/or higher total laser energy lenses or excimer laser photorefractive keratectomy. Am 1 required to produce a wide capsulotomy may also theo­ Ophthalmol 1993;115:444-53. retically carry an increased incidence of complications. 8. Van den Berg TJTP. Importance of pathological intraocular light scatter for visual disability. Doc Ophthalmol This theory is largely by clinical studies, unsubstantiated 1986;61:327. whereas the benefits in terms of reduction of forward light 9. Michelson AA. Studies in optics. University of Chicago scatter by capsule remnants are well domonstrated by this Press. Chicago. 1927. 1 . 6 study. 4 1 10. Pearce lL. Modernsimple extracapsular surgery. Trans Oph­ In conclusion, in order to minimise forward scatter thalmol Soc UK 1979;99: 170-6. II. Wilhelmus KR, Emery 1M. Posterior capsule opacification from capsule remnants and reduce the consequent glare following . Ophthalmol Surg 1980: II: disability of the patient, it is recommended that pseudo­ 264-7. phakic patients with posterior capsular opacification have 12. Nadler IN. Glare and contrast sensitivity in cataracts and their pupils dilated during Nd:YAG laser capsulotomy so pseudophakia. In: Nadler MP. Miller D, Nadler Dl, editors. that a wide capsulotomy can be performed. If laser treat­ Glare and contrast sensitivity for clinicians. Berlin: Springer, 1990. ment is undertaken simply to improve Snellen acuity, a 13. Maltzman BA, Horan C, Rengal A. Penlight test for glare small capsulotomy appears to be as effective as a wide disability of cataracts. Ophthalmic Surg 1988;19:356-8. capsulotomy. 14. Steinert RF, Puliafito CA. Kumar SR, Dudak SD, Patel S. Cystoid macular edema, retinal detachment. and glaucoma Key words: Caps ulotomy. Glare. Light scatter. after Nd:YAG laser posterior capsulotomy. Am 1 Ophthal­ mol 1991;112:373-80. REFERENCES 15. Holladay 1. Bishop 1. Lewis 1. The optimal size of a capsulo­ I. Courtney P. The National Survey. I. tomy. Am Intraocul Implant Soc 1 1985; II:18-20. Method and descriptive features. Eye 1992;6:487-92. 16. Bukelman A. Abrahami S, Oliver M. Pollack A. YAG cap­ 2. Hodgkins PRo Luff Al. Morrell Al. Teye Botchway L. sulotomy and cystoid macular edema. Eye 1992;6:35-8. Featherstone Tl, Fielder A. Current practice of extraction 17. Lewis H, Singer TR, Hanscom TA, Straasma BR. A pro­ and anaesthesia. Br 1 Ophthalmol 1992;76:323--6. spective study of cystoid macular edema after neody ­ 3. Pizzarello LD. The dimensions of the problem of eye mium:Y AG laSEr posterior capsulotomy. Ophthalmology disease among the elderly. Ophthalmology 1987;94: 1987;94:478-82. 1191-5. 18. MacEwen Cl. Baines PS. Retinal detachment following 4. Aron-Rosa DS, Aron 11, Cohn H. Use of a pulsed pico- YAG laser capsulotomy. Ey e 1989;3:759-63.