Visual Performance with Multifocal Intraocular Lenses

Transcription

1 Visual Performance with Multifocal Intraocular Lenses Mesopic Contrast Sensitivity under Distance and Near Conditions Robert Montés-Micó, OD, MPhil, 1 Enrique España, MD, PhD, 1,2 Inmaculada Bueno, OD, 1 W. Neil Charman, PhD, DSc, 3 José L. Menezo, MD, PhD 2,4,5 Objective: To evaluate distance and near visual performance under bright (photopic) and dim (mesopic) conditions in patients who had undergone uncomplicated cataract extraction with multifocal or monofocal intraocular lens (IOL) implantation. Design: Prospective, nonrandomized, masked, comparative, observational case series. Participants: Thirty-two eyes of 32 patients after zonal-progressive multifocal IOL implantation (Allergan Medical Optics Array SA-40N) and 32 eyes of 32 age-matched patients after monofocal IOL implantation (Allergan Medical Optics SI-40NB). Intervention: All eyes underwent phacoemulsification and IOL implantation. Main Outcome Measures: At 18 months after surgery, the monocular contrast sensitivity (CS) function was measured with sinusoidal grating charts at distance and near, at one photopic luminance level and 2 mesopic luminance levels (85, 5, and 2.5 candelas per square meter). Results: Under bright conditions, CS at distance in the multifocal group was not statistically different (P 0.01) from that in the monofocal group at any tested grating spatial frequency (1.5, 3, 6, 12, and 18 cycles per degree [cpd]). At low luminances, distance CS for the multifocal group was worse than that for the monofocal group at the highest test spatial frequencies (12 and 18 cpd; P 0.01). At near, photopic CS in the multifocal group was lower than at distance; patients with only a monofocal distance correction, however, could not detect the test gratings, even at the highest available contrast. With optimal near spectacle additions (i.e., using the distance correction of the multifocal IOL), there were no significant differences between the photopic near CS values for the multifocal and monofocal groups. When the luminance was decreased, near CS at all spatial frequencies was reduced in both groups. Contrast sensitivity in the near-corrected, multifocal group was significantly worse than in the near-corrected, monofocal group at high spatial frequencies (12 and 18 cpd). Conclusions: This work supports the findings of earlier authors that the Array multifocal IOL, with its center-distance design, is distance biased. Distance CS is within normal limits under bright photopic conditions but shows deficits at higher spatial frequencies (more than approximately 12 cpd) under dim mesopic conditions. Near CS obtained with the multifocal IOL is below that which can be achieved by an appropriate monofocal near correction, for all spatial frequencies and illumination conditions. Ophthalmology 2004;111: by the American Academy of Ophthalmology. The visual performance of patients who have undergone cataract extraction depends on the type of intraocular lens Originally received: July 12, Accepted: May 21, Manuscript no Optometry and Vision Sciences Unit, University of Valencia, Valencia, Spain. 2 Department of Ophthalmology, La Fe University Hospital, Valencia, Spain. 3 Department of Optometry and Neuroscience, University of Manchester Institute of Science and Technology, Manchester, United Kingdom. 4 Department of Surgery, University of Valencia, Valencia, Spain. 5 Mediterranean Ophthalmological Foundation, Valencia, Spain. The authors have no proprietary interest in any of the materials mentioned in this article. (IOL) that has been implanted. Monofocal IOLs provide excellent visual function, but, for many patients, their limited depth of focus means that they cannot provide clear vision at both distance and near, although this may be achieved in some patients particularly in those with small pupils. 1,2 Monovision techniques may be helpful for some patients but involve some sacrifices in binocularity. In general, if the monofocal IOL power is selected for distance correction, reading spectacles will be required. Recently, multifocal IOLs have been developed that offer the pseu- Correspondence to Robert Montés-Micó, OD, MPhil, Optometry and Vision Sciences Unit, Optics Department, Physics Faculty, University of Valencia, C/Dr. Moliner, , Burjassot, Valencia, Spain. roberto.montes@uv.es by the American Academy of Ophthalmology ISSN /04/$ see front matter Published by Elsevier Inc. doi: /s (03)

2 Ophthalmology Volume 111, Number 1, January 2004 dophakic patient the possibility of satisfactory vision at both distance and near conditions without the use of spectacles These simultaneous-vision lenses provide distance, intermediate, and near correction within the area of the ocular pupil. When a distant object is viewed, a sharp retinal image is provided by those parts of the lens within the pupillary area that have the distance correction, and a somewhat blurred image is provided by the other parts of the lens these images are superimposed on the retina. The roles of the corrections change when a near object is observed; then, those regions of the lens occupied by the near correction provide the correctly focused retinal image. In both situations, the unwanted effect of the light in the out-of-focus image is to reduce the contrast of the in-focus image. 12,13 Visual quality can, however, still be fully acceptable to the patient, and there may be little effect on high-contrast distance and near acuities. Approaches to multifocal IOLs can be broadly categorized as diffractive or refractive. Diffractive multifocal IOLs were the first to be evaluated clinically, but they showed significant optical deficiencies that discouraged many ophthalmologists from using them. In an attempt to avoid the problems of diffractive multifocal IOLs, refractive, bifocal, and zonal-progressive multifocal lenses were developed. Despite an improvement in uncorrected near visual acuity with these lenses, loss of clarity, loss of low-contrast acuity, and complaints of halos and glare have been reported with some types of lens, and further potential exists for design improvements. The visual performance of patients implanted with refractive multifocal IOLs depends on 2 important factors: the characteristics of the lens (i.e., the optical design of its refractive surfaces) and the visual conditions for the patient (i.e., distance or near vision and the illumination level, because the latter affects both the ocular pupil size and neural performance). The skills of the surgeon in selecting and positioning the lens to give optimal power and centration, combined with minimal astigmatism, are obviously also important. If current multifocal designs are to be refined, it is important that the interaction between lens design and visual performance under different conditions be understood as fully as possible. One of the more popular current refractive multifocal lenses is the Allergan Medical Optics (AMO) Array multifocal IOL (Allergan, Irvine, CA). This was the first multifocal IOL to be approved by the U.S. Food and Drug Administration and, in view of its widespread use, was chosen for this study. The foldable AMO Array multifocal IOL has a series of repeatable, continual aspheric power distributions on the anterior surface of the lens, shown schematically in Figure 1. The power profile was designed to smooth the transitions between zones, diminishing the appearance of halos around light and allowing for a range of foci from far to near. As can be seen in Figure 1, the lens optic area has a diameter of 4.7 mm, divided into 5 main annular zones. The central, circular zone has the power required for the distance correction, and then there are 3 main annular zones of increasing diameter containing, in sequence, primarily near, distance, and, again, near power; the zone diameters are as given in Figure 1. The maximum Figure 1. Schematic representation of the zonal-progressive multifocal lens design of the Allergan Medical Optics Array multifocal intraocular lens. D diopters. add power is 3.5 diopters (D), corresponding to approximately 2.8 D in the spectacle plane. 9 There are narrower blending zones of varying power between the main zones, and the fifth annular zone ( mm diameter) is a transitional zone to the distance correction in the lens periphery. The zonal design of the lens makes it relatively forgiving to small amounts of decentration and tilt. It seems reasonable to suppose that the visual performance obtained with multifocal lenses of this general type will be a function of pupil size, because this will affect the relative areas of the pupil occupied by the near and distance corrections. However, some controversy still exists in relation to the suggestion that pupil size might affect the visual response with multifocal IOLs. Ravalico et al 22 found no differences in visual acuity as a result of changes in pupillary size. In contrast, a recent study of Hayashi et al 23 concluded that pupil size does influence visual acuity in eyes with multifocal IOLs. The major factor affecting pupil size is the illumination level. Montés-Micó and Charman 24 found that, relative to normal emmetropes, patients who had undergone excimer laser photorefractive keratectomy showed significant decreases in contrast sensitivity (CS) under mesopic conditions. They attributed this to increased postsurgical optical aberrations for the larger pupil sizes found under mesopic conditions. It may be expected that illumination-dependent changes in pupil diameter would also affect the quality of vision achieved by patients with multifocal IOLs. Because these lenses are designed to provide good vision at distance and near conditions, the visual response needs to be evaluated under both conditions. There have been few reports of an assessment of visual performance with multifocal IOLs as a function of the illumination level. Because the pupil normally shows some constriction when near objects are observed, as part of the action of the near triad, 25 a center-distance multifocal, such as that 86

3 Montés-Micó et al Mesopic Contrast Sensitivity Function in Multifocal IOL used in this study, might be expected to provide somewhat compromised near vision. The performance index that most usefully documents human spatial vision is the CS function (CSF), which plots the reciprocal of the threshold contrast for sinusoidal gratings as a function of their spatial frequency. It thus gives information on visual performance for a range of object scales and is especially useful in patients who have undergone refractive surgery procedures. 26 Thus, the objective of this study was to study the CSF, as a function of the illumination level under distance and near conditions, of patients implanted with AMO Array multifocal IOLs and to compare the visual performance found with that achieved by a similar group of patients implanted with monofocal IOLs. Materials and Methods We prospectively examined 32 eyes of 32 consecutive patients who underwent implantation of the AMO Array SA-40N multifocal IOL and 32 eyes of 32 age-matched patients who underwent implantation of the monofocal AMO SI-40NB IOL. For each patient with a multifocal IOL, a control patient with a monofocal IOL was selected who was within 5 years of the same age. Exclusion criteria for both groups included ocular disease other than cataract, history of ocular surgery or inflammation, and astigmatism 1.50 D. The amount of cylinder was limited to 1.75 D, because larger amounts of cylinder have been shown to decrease near vision capabilities with the Array lens design. 8 All cataracts in this study were extracted by the same surgeon (JLM) by using topical anesthesia and a clear corneal 3.2-mm temporal incision. Phacoemulsification was followed by irrigation and aspiration of the cortex and IOL implantation in the capsular bag. All patients were happy with the outcome of their surgery. The AMO Array SA-40N multifocal IOL is structurally identical to the AMO SI-40NB monofocal IOL except for the lens optic s multifocal design. The overall diameter of the lenses was 13.0 mm, and the optical diameter was 6.0 mm. Lens power varied from to D, and, in the case of the AMO Array multifocal IOL, the near power increased by 3.50 D (Fig 1). The tenets of the Declaration of Helsinki were followed in this research. Informed consent was obtained from all patients after the nature and possible consequences of the study had been explained. The CS of both groups was measured with the Stereo Optical Functional Acuity Contrast Test (Stereo Optical, Chicago, IL). This allows presentation of sine-wave gratings of different spatial frequencies (1.5, 3, 6, 12, and 18 cycles per degree [cpd]), with contrasts changing in steps corresponding to 0.15 log 10 CS. The manufacturer s recommended testing procedures were followed; the testing distance was 3 m for distance and 40 cm for near vision. Absolute values of log 10 CS were obtained for each combination of patient, spatial frequency, luminance, and distance and near vision, and means and standard deviations were calculated. Measurements were obtained monocularly on 32 eyes for the control group and on 32 eyes for the multifocal group 18 months ( 5 months) after IOL implantation. The nonviewing eye (left eye) was occluded for each measurement, and no refractive correction was initially used with the viewing eye (right eye), in accordance with the normal practice of the individuals concerned. However, to determine whether residual refractive error (defocus) might be modifying the CS, the measurement procedure was repeated with appropriate additional optimal spectacle corrections for distance and near vision. The CS measurements were taken under 3 different illumination conditions. The levels of chart luminance were 85, 5, and 2.5 candelas per square meter (cd/m 2 ). The first was photopic (i.e., the luminance recommended in the manufacturer s guidelines), and the others were mesopic; room illumination was at similar levels. The luminance levels were achieved by means of variable diffuse illumination. Contrast sensitivity was measured first at the photopic level and then under the mesopic levels. Patients adapted to each level for 5 minutes before testing. As noted previously, visual performance was expected to be dependent on pupil size. Pupil diameters in distance vision were therefore measured in each patient under the different levels of illumination by means of a pupillometer (Colvard; Oasis, Glendora, CA). All examinations were performed by 2 ophthalmic technicians masked as to the IOL status of each eye. Results Table 1. Demographic Characteristics of Participants Variable Multifocal IOL Monofocal IOL P Value No. of eyes Age (yrs) Gender (M/F) 18/14 19/ Astigmatism (diopters)* Interval (mos) Pupil diameter (mm) Luminance level (photopic) Luminance level (mesopic) Luminance level 3 (mesopic) IOL intraocular lens. *Postoperative corneal astigmatism. Elapsed time between surgery and examination. Level 1, 85 candelas per square meter (cd/m 2 ); level 2, 5 cd/m 2 ; level 3, 2.5 cd/m 2. Patient demographics and pupil diameters in distance vision are shown in Table 1; pupils were slightly smaller during near vision. There were no statistically significant differences between the multifocal and monofocal IOL groups with respect to age, gender, corneal astigmatism, pupil diameter at different illumination conditions, and interval between surgery and evaluation. The normality of data distribution was tested by using the Kolmogorov Smirnov test. The means and standard deviations of log 10 CS as a function of spatial frequency under the various lens and illumination conditions are given for the distance tests in Table 2 and for the near tests in Table 3. To explore the statistical significance of differences between the multifocal and monofocal groups, t tests were performed on the comparable data of the 2 groups (absolute log 10 CS values) at each spatial frequency and illumination level. The results are shown in Tables 4 and 5 for distance and near vision, respectively. Any differences showing a P value of 0.01 (i.e., at the 1% level) were considered to be statistically significant. Although it could be argued that a lower P value might be more appropriate in view of the large number of comparisons performed (n 90), this level was believed to provide a reasonable indication of those combinations of parameters for which real differences in visual performance occurred. In fact, some 42 comparisons 87

4 Ophthalmology Volume 111, Number 1, January 2004 Table 2. Distance Log 10 Contrast Sensitivity Values at Three Luminance Levels (85, 5, and 2.5 Candelas per Square Meter [cd/m 2 ]) for Each Intraocular Lens Condition Spatial Frequency (cpd) MoU MoC MU MC 85 cd/m (0.16) 1.89 (0.15) 1.73 (0.11) 1.82 (0.13) (0.13) 2.16 (0.10) 2.02 (0.11) 2.08 (0.12) (0.17) 2.09 (0.11) 1.94 (0.12) 2.01 (0.10) (0.09) 1.98 (0.06) 1.85 (0.10) 1.81 (0.13) (0.10) 1.48 (0.06) 1.38 (0.10) 1.42 (0.10) 5 cd/m (0.10) 1.77 (0.09) 1.57 (0.12) 1.64 (0.20) (0.12) 1.92 (0.15) 1.72 (0.14) 1.86 (0.13) (0.14) 1.69 (0.13) 1.55 (0.16) 1.69 (0.11) (0.11) 1.35 (0.07) 0.85 (0.09) 0.89 (0.12) (0.08) 1.01 (0.09) 0.59 (0.08) 0.62 (0.09) 2.5 cd/m (0.14) 1.71 (0.08) 1.49 (0.14) 1.55 (0.14) (0.18) 1.80 (0.14) 1.60 (0.18) 1.74 (0.17) (0.15) 1.59 (0.08) 1.36 (0.15) 1.46 (0.14) (0.09) 1.11 (0.07) 0.72 (0.07) 0.68 (0.08) (0.06) 0.83 (0.08) 0.48 (0.08) 0.49 (0.09) cpd cycles per degree; MC multifocal best corrected for distance; MoC monofocal best corrected for distance; MoU monofocal uncorrected; MU multifocal uncorrected. Values are means for each patient group (standard deviation). showed P 0.01 far more than the 4 or 5 that might be expected to arise by chance alone in a large number of tests of this type. The mean values of log 10 CS are plotted as a series of CSFs in Figure 2. Figure 2a c shows distance CSFs at the 3 luminance levels, and Figure 2d f shows the results of the measurements at near conditions. Standard adult mean measurements for normal eyes as found by Boxer Wachler and Krueger 27 for the Functional Acuity Contrast Test chart at a luminance level of 85 cd/m 2 are included in each panel for comparison. At distance, under photopic conditions (85 cd/m 2 ), performance was very similar for all groups and was close to that found by Boxer Wachler and Krueger 27 (Fig 2a). At the mesopic levels of 5 and 2.5 cd/m 2, however, CS was generally lower, particularly at higher spatial frequencies (Fig 2b, c). This behavior corresponds almost exactly to data found for healthy emmetropic eyes under similar testing conditions by Montés-Micó and Charman. 24 Note that the data for uncorrected multifocal and monofocal eyes in Figure 2a c show slightly lower values of normalized CS than when a suitable additional distance correction is used, presumably because of small amounts of residual refractive error and also, perhaps, because the distance CS test was performed at 3 m, rather than 6 m. Mesopic log 10 CS at 12 and 18 cpd was significantly lower in the multifocal cases than in their monofocal counterparts (P 0.01; Table 4). Table 3. Near Log 10 Contrast Sensitivity Values at Three Luminance Levels (85, 5, and 2.5 Candelas per Square Meter [cd/m 2 ]) for Each Intraocular Lens Condition Spatial Frequency (cpd) MoNC MU MC MNC 85 cd/m (0.12) 1.35 (0.10) 1.27 (0.10) 1.82 (0.09) (0.10) 1.54 (0.09) 1.49 (0.10) 2.09 (0.13) (0.14) 1.61 (0.12) 1.69 (0.13) 2.14 (0.11) (0.11) 1.44 (0.10) 1.48 (0.11) 1.91 (0.14) (0.10) 1.10 (0.11) 1.13 (0.10) 1.54 (0.08) 5 cd/m (0.09) 0.89 (0.08) 0.91 (0.09) 1.37 (0.12) (0.10) 0.89 (0.09) 0.91 (0.10) 1.41 (0.13) (0.11) 0.77 (0.10) 0.88 (0.11) 1.36 (0.14) (0.11) 0.59 (0.10) 0.65 (0.11) 1.05 (0.08) (0.08) 0.42 (0.08) 0.44 (0.07) 0.78 (0.07) 2.5 cd/m (0.10) 0.80 (0.11) 0.84 (0.12) 1.27 (0.10) (0.11) 0.79 (0.13) 0.75 (0.13) 1.31 (0.11) (0.13) 0.69 (0.15) 0.65 (0.08) 1.19 (0.15) (0.13) 0.52 (0.15) 0.56 (0.14) 0.96 (0.13) (0.09) 0.36 (0.08) 0.42 (0.10) 0.71 (0.08) cpd cycles per degree; MC multifocal best corrected for distance (i.e., using near addition of multifocal); MNC multifocal best corrected for distance with near addition (i.e., using the distance power of the multifocal); MoNC monofocal near corrected; MU multifocal uncorrected. Values are means for each patient group (standard deviation). 88

5 Montés-Micó et al Mesopic Contrast Sensitivity Function in Multifocal IOL Table 4. Significance Levels (P Values) for the Differences in Log 10 Distance Contrast Sensitivity between Multifocal and Monofocal Groups at Each Spatial Frequency and Luminance Luminance Level (cd/m 2 ) Spatial Frequency (cpd) MU/MoU MC/MoC MU/MoU MC/MoC MU/MoU MC/MoC MU/MoU * * MC/MoC * * 18 MU/MoU * * MC/MoC * * cd candelas; cpd cycles per degree; MC multifocal best-corrected distance; MoC monofocal best-corrected distance; MoU monofocal uncorrected; MU multifocal uncorrected. *Statistically significant. At near, CS evaluation of the monofocal group, either with or without a distance correction, was not possible because the patients could not detect the gratings at the maximum available chart contrast at any spatial frequency. Only upper limits to normalized CS could be determined. This was obviously due to the limited depth of focus obtained with the monofocal correction and the absence of accommodation. 28 The photopic CSF for the monofocal patients with an optimal near addition was very similar to that found at distance (Fig 2d). For the multifocal patients relying on the near addition provided by the IOL, performance was significantly worse than in the near-corrected monofocal group (P 0.01; Table 5), even when any residual distance error for the multifocal IOL was corrected. If, however, the multifocal patients were given an extra near addition, so that they were effectively using the distance correction provided by the IOL to view the near targets, the CSF approached normal levels (Fig 2d). Under dim, mesopic conditions (2.5 and 5 cd/m 2 ; Fig 2e, f), as in the distance case, near CS was always lower than that achieved under photopic conditions. Near CS with both the uncorrected multifocal IOL and the distance-corrected multifocal IOL i.e., when patients relied on the near addition provided by the multifocal IOL to view the near gratings was always significantly worse than that achieved with the monofocal IOL with a near addition (P 0.01; Table 5). Only when an additional spectacle near addition was used with the IOL, so that the patients again relied on the distance correction provided by the IOL to view the near gratings, did the multifocal patients achieve performance comparable to that of the near-corrected monofocal patients. This implies that the distance image provided by the Array multifocal is of superior optical quality to the near image. 9 Note that, even with the extra near addition, mesopic multifocal performance was still significantly worse than that with the monofocal and near addition at 12 and 18 cpd (Table 5). Discussion Previous multinational clinical trials evaluating clinical, functional, and quality of life outcomes after AMO Array Table 5. Significance Levels (P Values) for the Differences in Log Near Contrast Sensitivity between Multifocal and Monofocal Groups at Each Spatial Frequency and Luminance Luminance Level (cd/m 2 ) Spatial Frequency (cpd) MoNC/MNC MC/MU MU/MNC * * * MoNC/MC * * * 3 MoNC/MNC MC/MU MU/MNC * * * MoNC/MC * * * 6 MoNC/MNC MC/MU MU/MNC * * * MoNC/MC * * * 12 MoNC/MNC * * MC/MU MU/MNC * * * MoNC/MC * * * 18 MoNC/MNC * * MC/MU MU/MNC * * * MoNC/MC * * * cd candelas; cpd cycles per degree; MC multifocal best-corrected distance; MNC multifocal near corrected; MoNC monofocal near corrected; MU multifocal uncorrected. *Statistically significant. IOL implantation 8 11 have shown that this IOL can improve near vision while providing a good level of distance vision. Patients with multifocal IOLs reported less limitation in visual function and less spectacle dependency than patients with bilateral monofocal IOLs. Effect of Pupil Diameter Before we discuss the CSF results in more detail, it is helpful to consider the optical effects of luminance-dependent changes in pupil diameter with the multifocal IOL. If the power profile in Figure 1 is considered, it is logical to expect some dependence of the retinal image and, hence, visual performance, on pupil diameter. For example, with a well-centered lens, a small, 2.1-mm pupil will nominally receive only a distance correction: there will be no effective near correction, although depth of focus will be reasonable, because of the pinhole effect. In fact, the corresponding entrance pupil of the eye will be somewhat larger than this approximately 2.4 mm in diameter because of the magnifying effects of the cornea (a simple schematic eye with a corneal radius of curvature of 8 mm, an anterior chamber depth of 4 mm, and an aqueous humor refractive index of has been assumed, yielding a magnification of 1.15 between the anterior surface of the lens and the entrance pupil of the eye). As the pupil dilates, progressively more of the pupil receives a near correction. Using 89

6 Ophthalmology Volume 111, Number 1, January 2004 Figure 2. Contrast sensitivity (CS) functions for the different intraocular lens groups as a function of test distance and Functional Acuity Contrast Test chart luminance: (a c) distance conditions and (d e) near conditions. For comparison, the standard data represented by the dashed lines are photopic (85 candelas per square meter [cd/m 2 ]) results for normal, healthy emmetropic eyes as found by Boxer Wachler and Krueger. 27 Luminance levels: a, d, 85 cd/m 2 ; b, e, 5 cd/m 2 ; c, f, 2.5 cd/m 2. cpd cycles per degree; MC multifocal best-corrected distance; MNC multifocal near corrected; MoC monofocal best-corrected distance; MoNC monofocal near corrected; MoU monofocal uncorrected; MU multifocal uncorrected. the zone dimensions given in Figure 1 and assuming the 1.15 magnification factor for the ocular entrance pupil, it is easy to calculate the proportion of the pupil that is occupied by the distance correction as a function of the diameter of the entrance pupil of the eye for a well-centered multifocal IOL. This is shown in Figure 3, where it can be seen that although for small pupils, the multifocal acts like a monofocal distance correction, for pupil diameters between approximately 3.5 and 5 mm (the mean range in this study; Table 1), approximately half the pupil area is occupied by the distance correction. This fraction remains roughly independent of pupil diameter within this range. 90

7 Figure 2. Continued. Light not contributing to an in-focus retinal image will reduce image contrast and, hence, effective CS. In practice, the blending of the refractive zones means that Figure 3 is unlikely to be an exact representation of the situation, but there can be no doubt that the ocular pupil diameter influences the balance of light that contributes to the images 91

8 Ophthalmology Volume 111, Number 1, January 2004 Figure 3. Approximate relative area of the pupil covered by the distance correction of a well-centered Array multifocal intraocular lens, as a function of the diameter of the entrance pupil of the eye. Other areas of the lens provide intermediate and near correction. formed by the distance and near regions of the lens and, hence, the contrast of the retinal image of any object. In general terms, too, larger pupils are associated with smaller depths of focus, because for any given error of focus, the out-of-focus retinal blur patches are of larger diameter. It may be that the lack of pupil dependence in the results of Ravalico et al 22 was due to the use of high-contrast letter charts to measure visual acuity: acuity measurements with such charts are little affected by modest reductions in contrast. 29 Photopic Results There have been several previous studies of photopic CS after multifocal IOL implantation 21,23,30 34 ; the most useful were those of Hayashi et al. 23,34 These researchers found that CS with a multifocal IOL was reduced compared with that for a monofocal IOL but was, however, within the normal range. Our results broadly agree with this. Contrast sensitivity with the multifocal correction tended to be slightly lower than with the monofocal correction, but the differences were not significant, and the mean multifocal CSF lay within normal limits. As discussed previously, decreased CS in patients with multifocal IOLs, as compared with patients with monofocal IOLs, can be explained by the multifocal IOL s division of the available light energy in the image between 2 or more focal points. Ravalico et al, 35,36 by means of a laser optical bench, found that approximately 50% of the light energy in the full-aperture AMO Array IOL was concentrated at the far focus and that approximately 20% was concentrated at the near focus; this is broadly compatible with the predictions of Figure 3. Because defocus primarily affects images at higher spatial frequencies, this would imply that, in comparison with an optimal monofocal correction, distance and near CS with the multifocal IOL might be lower at higher spatial frequencies by factors of up to approximately 0.5 (0.3 log units) and 0.2 (0.7 log units), respectively. However, clinical studies, including our own, show that the loss in photopic CS is somewhat smaller. It may be that the effects of ocular longitudinal chromatic and other types of ocular aberration, together with the blending zones of the IOL, tend to mask the differences between the monofocal and multifocal images and, hence, the differences in CS. Because no significant differences were obtained between the CSFs for the multifocal group with and without a distance correction at any spatial frequency and luminance level (distance and near), it seems unlikely that residual refractive errors were important (Fig 2). When the evaluation was performed at near, the CS of the multifocal group (both when uncorrected and best distance corrected) at the photopic level was worse than that obtained at distance (Fig 2a, d). The lower effectiveness of the near addition provided by the multifocal correction is to be expected. Figure 3 suggests that, for the pupil sizes in use, which were probably slightly smaller than those given in Table 1 because of pupillary constriction at near, only approximately half the light goes into the full range of near and intermediate foci. As noted previously, Ravalico et al 35,36 found that only approximately 20% of the light contributed to the near focus, in contrast to 50% in the far focus. Thus, the near retinal image is likely to be of reduced contrast in comparison with the distance image. A further factor is that the in-focus modulation transfer function of a lens with an annular aperture, as in the case of the near additions, is depressed at higher spatial frequencies in comparison to that with a circular pupil of the same outside diameter, 18 although the annular aperture does improve depth of focus. 40,41 Giving the multifocal group a near addition to view the near test charts brought the distance focus of the multifocal IOL onto the retina. Not surprisingly, the CSF achieved was then similar to that found in the distance tests. Our results showed an improvement of approximately 0.2 in logcs when the distance, as compared with the near, correction of the multifocal IOL was used. Sasaki 9 found similarly reduced values of CS at near conditions in relation to distance conditions. Effect of Dim Illumination on CS Several previous studies 9,32,33 have demonstrated that patients with multifocal IOLs show worse CS under dim 92

9 Montés-Micó et al Mesopic Contrast Sensitivity Function in Multifocal IOL Figure 4. Schematic diagram showing the general form of the point images at distance and near conditions for the Array lens on the simplifying assumption that it has a binary refractive distribution across its area (distance and near corrections only, with a 2.5-diopter near addition). The approximations of geometrical optics have been used to show the image blur for ocular entrance pupil diameters of 3.5, 3.9, and 4.6 mm, as observed at luminances of 85, 5, and 2.5 candelas per square meter (cd/m 2 ), respectively (Table 1). The images are shown as negatives; the dark spot at the center of each image is the in-focus image formed by the appropriate correction (distance for distance, near for near), and the surrounding circular or annular blur patches are formed by the inappropriate correction. conditions. Our results also show that, whatever the correction used, there is a reduction in CS at distance conditions when the luminance level is reduced, particularly at the higher spatial frequencies (Fig 2). This trend agrees with classic data about the effect of luminance level on CS. 42,43 Our photopic and mesopic IOL values agree quite well with the values of Montés-Micó and Charman 24 for healthy emmetropic phakic patients, except that the mesopic CS values of the multifocal group at frequencies of 12 and 18 cpd are significantly lower than those obtained in the monofocal group (Table 2; P 0.01). The pupil diameters are substantially larger under mesopic conditions (Table 1), and it therefore seems reasonable to attribute the observed reduction in multifocal mesopic CS at higher spatial frequencies, relative to the optimal monofocal correction, to the additional blur introduced by the larger-diameter, out-of-focus zones of the multifocal IOL. Whereas at photopic levels, only zones 1 and 2 (Fig 1) will typically contribute to the retinal images, at the lower light levels, zones 3 and 4 will also be involved. For any level of defocus, the size of the annular blur patches will scale with the dimensions of the zone of the lens. This is shown schematically in Figure 4, which indicates the basic form of the retinal point images at the 3 illumination levels. It has been assumed that the multifocal IOL has a binary refractive profile (i.e., that its local power is appropriate for either distance or near vision, with the zone diameters given in Figure 1, rather than sometimes being intermediate). Although the images achieved in practice will be more complex because of the blending of the zones, the effects of aberration, and small lens tilts and decentrations and because there is a range of pupil sizes among individual patients at any luminance level (Table 1), the increased blur associated with the larger pupil, as shown in Figure 4, is likely to be broadly correct. At this stage, it is helpful to try to summarize the relative merits of the monofocal and multifocal corrections. How does the vision, at distance and near conditions, of a patient with a multifocal IOL compare with an individual equipped with a monofocal IOL who wears a spectacle correction for near vision? We can make this comparison at the 3 luminance levels used in the study by dividing, at each spatial frequency, the CS obtained with the multifocal IOL by that obtained with the monofocal IOL, with the latter being used alone at distance conditions and with additional correction at near conditions. To avoid any problems caused by residual uncorrected distance refractive errors, it seems reasonable to use each basic IOL with any required distance correction. Figure 5 gives the results. Standard deviations 93

10 Figure 5. Contrast sensitivity (CS) with the distance-corrected multifocal patients expressed in terms of the CS for the monofocal group. For distance vision, the latter wore a distance correction for any residual refractive error, and for near vision, they wore a near addition. Ordinates show the CS of the multifocal group divided by that for the monofocal group. Diamonds (solid line) show data for distance vision; squares (dashed line) show data for near vision. Luminances were (a) 85 candelas per square meter (cd/m 2 ), (b) 5 cd/m 2, and (c) 2.5 cd/m 2. 94

11 Montés-Micó et al Mesopic Contrast Sensitivity Function in Multifocal IOL are omitted for clarity but typically amount to approximately 30% of the value of each data point. When presented in this way, the data again show that, under photopic conditions, multifocal CS at distance is only slightly worse than that with a monofocal IOL. Near CS for the multifocal group was, however, only approximately 30% of that achieved by the monofocal patients when they were provided with a near spectacle correction (i.e., 0.5 log units worse). At the 2 dim mesopic levels, the decrease in relative distance CS with the multifocal at the higher 12- and 18-cpd frequencies was obvious, whereas near CS with the multifocal correction was again much worse than that achievable with a monofocal IOL and a near addition. Hence, in exchange for the convenience of not using reading spectacles, the patient with the multifocal IOL may encounter some problems with distance vision at low luminances and with near vision for lower-contrast objects. To summarize, the AMO Array multifocal IOL provides good CS at a distance under photopic conditions; the performance is comparable to that obtained with monofocal IOLs. At near, the photopic CSF is lower than at distance, although it is acceptable for the everyday needs of many patients. Contrast sensitivity function at near can be somewhat improved by use of a near addition, but this negates the purpose for which the multifocal IOL was designed. Those with critical visual needs at near conditions would be better served with a monofocal correction and a spectacle near correction. Under mesopic conditions, distance CS at high spatial frequencies with the Array IOL is significantly worse that that with a monofocal correction. This last consideration should be kept in mind when advising patients who need optimal spatial vision under low luminance or night illumination conditions, as might be encountered when driving at night. References 1. Tucker J, Rabie E. Depth-of-focus of the pseudophakic eye. Br J Physiol Opt 1980;34: Nakazawa N, Ohtsuki K. Apparent accommodation in pseudophakic eyes after implantation of posterior chamber intraocular lenses. Am J Ophthalmol 1983;96: Hansen TE, Corydon L, Krag S, Thim K. New multifocal intraocular lens design. J Cataract Refract Surg 1990;16: Duffey RJ, Zabel RW, Lindstrom RL. Multifocal intraocular lenses. J Cataract Refract Surg 1990;16: Menezo JL, Ferrer E, Taboada JF, Cisneros AL. Nuestros resultados en el estudio mundial multicéntrico de L.I. O multifocales. Arch Soc Esp Oftalmol 1990;59: Taboada JF, Menezo JL, Ferrer E. Complicaciones de las L.I.O bifocales y multifocales [Bifocal and multifocal IOLs complication]. Arch Soc Esp Oftalmol 1991;61: Lindstrom RL. Food and Drug Administration study update. One-year results from 671 patients with the 3M multifocal intraocular lens. Ophthalmology 1993;100: Steinert RF, Aker BL, Trentacost DJ, et al. A prospective comparative study of the AMO ARRAY zonal-progressive multifocal silicone intraocular lens and a monofocal intraocular lens. Ophthalmology 1999;106: Sasaki A. Initial experience with a refractive multifocal intraocular lens in a Japanese population. J Cataract Refract Surg 2000;26: Javitt J, Brauweiler HP, Jacobi KW, et al. Cataract extraction with multifocal intraocular lens implantation: clinical, functional, and quality-of-life outcomes. Multicenter clinical trial in Germany and Austria. J Cataract Refract Surg 2000;26: Javitt JC, Steinert RF. Cataract extraction with multifocal intraocular lens implantation: a multinational clinical trial evaluating clinical, functional, and quality-of-life outcomes. Ophthalmology 2000;107: Navarro R, Ferro M, Artal P, Miranda I. Modulation transfer functions of eyes implanted with intraocular lenses. Appl Opt 1993;32: Artal P, Marcos S, Navarro R, et al. Through focus image quality of eyes implanted with monofocal and multifocal intraocular lenses. Opt Eng 1995;34: Ellingson FT. Explanation of 3M diffractive intraocular lenses. J Cataract Refract Surg 1990;16: Jay JL, Chakrabarti HS, Morrison JD. Quality of vision through diffractive bifocal intraocular lens. Br J Ophthalmol 1991;75: Gimbel HV, Sanders DR, Raanan MG. Visual and refractive results of multifocal intraocular lenses. Ophthalmology 1991; 98:881 7, discussion Post CT Jr. Comparison of depth of focus and low-contrast acuities for monofocal versus multifocal intraocular lens patients at 1 year. Ophthalmology 1992;99: , discussion Charman WN, Murray IJ, Nacer M, O Donoghue EP. Theoretical and practical performance of a concentric bifocal implant lens. Vision Res 1998;38: Pieh S, Lackner B, Hanselmayer G, et al. Halo size under distance and near conditions in refractive multifocal intraocular lenses. Br J Ophthalmol 2001;85: Schmitz S, Dick HB, Krummenauer F, et al. Contrast sensitivity and glare disability by halogen light after monofocal and multifocal lens implantation. Br J Ophthalmol 2000;84: Holladay JT, van Dijk H, Lang A, et al. Optical performance of multifocal intraocular lenses. J Cataract Refract Surg 1990; 16: Ravalico G, Baccara F, Bellavitis A. Refractive bifocal intraocular lens and pupillary diameter. J Cataract Refract Surg 1992;18: Hayashi K, Hayashi H, Nakao F, Hayashi F. Correlation between pupillary size and intraocular lens decentration and visual acuity of a zonal-progressive multifocal lens and a monofocal lens. Ophthalmology 2001;108: Montés-Micó R, Charman WN. Mesopic contrast sensitivity function after excimer laser photorefractive keratectomy. J Refract Surg 2002;18: Alpern M, Mason GL, Jardinico RE. Vergence and accommodation V. Pupil size changes associated with changes in accommodative convergence. Am J Ophthalmol 1961;52: Montés-Micó R, Charman WN. Choice of spatial frequency for contrast sensitivity evaluation after corneal refractive surgery. J Refract Surg 2001;17: Boxer Wachler BS, Krueger RR. Normalized contrast sensitivity values. J Refract Surg 1998;14: Bueno I, España E, Montés-Micó R, et al. Insufficiency of accommodation in patients with advanced AIDS. Ann Ophthalmol 2002;34: Regan D, Neima D. Low-contrast letter charts as a test of visual function. Ophthalmology 1983;90:

12 Ophthalmology Volume 111, Number 1, January Akutsu H, Legge GE, Showalter M, et al. Contrast sensitivity and reading through multifocal intraocular lenses. Arch Ophthalmol 1992;110: Williamson W, Poirier L, Coulon P, Verin P. Compared optical performances of multifocal and monofocal intraocular lenses (contrast sensitivity and dynamic visual acuity). Br J Ophthalmol 1994;78: Pieh S, Weghaupt H, Skorpik C. Contrast sensitivity and glare disability with diffractive and refractive multifocal intraocular lenses. J Cataract Refract Surg 1998;24: Yang HC, Chung SK, Baek NH. Decentration, tilt, and near vision of the array multifocal intraocular lens. J Cataract Refract Surg 2000;26: Hayashi K, Hayashi H, Nakao F, Hayashi F. Influence of astigmatism on multifocal and monofocal intraocular lenses. Am J Ophthalmol 2000;130: Ravalico G, Parentin F, Baccara F. Effect of astigmatism on multifocal intraocular lenses. J Cataract Refract Surg 1999; 25: Ravalico G, Parentin F, Sirotti P, Baccara F. Analysis of light energy distribution by multifocal intraocular lenses through an experimental optical model. J Cataract Refract Surg 1998;24: Green DG, Campbell FW. Effect of focus on the visual response to a sinusoidally modulated spatial stimulus. J Opt Soc Am 1965;55: Charman WN. Effect of refractive error in visual tests with sinusoidal gratings. Br J Physiol Opt 1979;33: Legge GE, Mullen KT, Woo GC, Campbell FW. Tolerance to visual defocus. J Opt Soc Am A 1987;4: Linfoot EH, Wolf E. Diffraction images in systems with an annular aperture. Proc Phys Soc 1953;B66: Welford WT. Use of annular apertures to increase focal depth. J Opt Soc Am 1960;50: Van Nes FL, Bouman MA. Spatial modulation transfer in the human eye. J Opt Soc Am 1967;57: Albarrán-Diego C, Montés-Micó R, Pons A, Artigas JM. Influence of the luminance level on visual performance with a disposable soft cosmetic tinted contact lens. Ophthalmic Physiol Opt 2001;21: