Share this article on:

Optical Performance of Multifocal Soft Contact Lenses via a Single-Pass Method

Bakaraju, Ravi C.; Ehrmann, Klaus; Falk, Darrin; Ho, Arthur; Papas, Eric

doi: 10.1097/OPX.0b013e318264f3e9
Original Articles

Purpose. A physical model eye capable of carrying soft contact lenses (CLs) was used as a platform to evaluate optical performance of several commercial multifocals (MFCLs) with high- and low-add powers and a single-vision control.

Methods. Optical performance was evaluated at three pupil sizes, six target vergences, and five CL-correcting positions using a spatially filtered monochromatic (632.8 nm) light source. The various target vergences were achieved by using negative trial lenses. A photosensor in the retinal plane recorded the image point-spread that enabled the computation of visual Strehl ratios. The centration of CLs was monitored by an additional integrated en face camera. Hydration of the correcting lens was maintained using a humidity chamber and repeated instillations of rewetting saline drops.

Results. All the MFCLs reduced performance for distance but considerably improved performance along the range of distance to near target vergences, relative to the single-vision CL. Performance was dependent on add power, design, pupil, and centration of the correcting CLs. Proclear (D) design produced good performance for intermediate vision, whereas Proclear (N) design performed well at near vision (p < 0.05). AirOptix design exhibited good performance for distance and intermediate vision. PureVision design showed improved performance across the test vergences, but only for pupils ≥4 mm in diameter. Performance of Acuvue bifocal was comparable with other MFCLs, but only for pupils >4 mm in diameter. Acuvue Oasys bifocal produced performance comparable with single-vision CL for most vergences.

Conclusions. Direct measurement of single-pass images at the retinal plane of a physical model eye used in conjunction with various MFCLs is demonstrated. This method may have utility in evaluating the relative effectiveness of commercial and prototype designs.

*BOptom, PhD, FAAO

MEng, PhD


§MOptom, PhD, FAAO

MCOptom, PhD, FAAO

Brien Holden Vision Institute, Sydney, New South Wales, Australia, Vision Cooperative Research Centre, Sydney, New South Wales, Australia, and School of Optometry and Vision Sciences, University of New South Wales, Sydney, New South Wales, Australia.

Received September 26, 2011; accepted May 3, 2012.

Ravi C. Bakaraju University of New South Wales Level 5, Rupert Myers NW, Gate 14, Barker St Kensington, NSW 2033 Australia e-mail:

Every form of presbyopic ophthalmic correction has some inherent level of compromise because unlike the eye, they can only provide a “static” correction and do not change their power when objects at different distances are viewed.1 Until a way of restoring accommodation is discovered, many individuals will avail benefit from the presbyopic contact lens (CL) alternatives. The CL industry has invested many resources over the past years to produce several different presbyopic designs. Most of the designs are considered failed, and the failure is often attributed to the neural causes, or, in other words, to the adaptation of the visual system.212 Although many early presbyopes have been CL wearers for years, it is anticipated that only approximately 9% of them have ever tried bi-/multifocals.13,14 The current reality is that only a tiny proportion of presbyopic patient correction is by multifocal CLs, with their success rates remaining discouragingly low. Despite much scientific work already having been conducted in this field,1524 new research aimed at better understanding the factors that contribute toward an optimal presbyopic CL correction remains necessary and is of importance.

It is arguable that the optical characteristics of vision-correction devices, including CLs, can be conveniently assessed using computer ray tracing of theoretical model eyes.2528 However, criticism is often raised that such models cannot be readily used to test commercial lenses as precise geometrical information, generally unavailable for commercial CLs, is required for computer ray-tracing analyses. When measured or calculated geometries are used for the evaluation of CL performance, the associated uncertainties and assumptions compromise the accuracy and robustness of the method. The available techniques and instrumentation can only provide approximate input values for optical modeling, particularly for the intricate profiles of multifocal CLs (MFCLs). To address this issue, we developed a method of comparing the image quality at the retinal plane produced by soft CLs using an anatomically and optically equivalent physical model eye.29 This work presents the evaluation of in vitro optical performance of several commercially available simultaneous-vision MFCLs used in conjunction with the physical model eye. The aim of this study is to determine whether there are measurable differences between lens types or lens designs to confirm the hypothesis that certain designs demonstrate better optical performance characteristics under presbyopic conditions.

Back to Top | Article Outline


The physical model eye used in the experiment closely replicates both anatomical and optical properties of an average human eye, as described elsewhere.29 Using the translation function of its motor base, this physical model eye was configured to an unaccommodated state with a refractive error of −2.00 D, which was corrected one at a time with seven MFCLs and one single-vision CL. Details of the CLs are provided in Table 1.

Two CLs of each design, one each from different batches, were randomly selected for the evaluation process, and each CL was independently evaluated twice, yielding four measurements. The performance evaluation of the CLs in conjunction with the physical model eye was obtained using an optical bench setup (Fig. 1). Light from a 0.95 mW 632.8 nm HeNe laser source (JDS Uniphase, Milpitas, CA) is first attenuated with a neutral density filter (ND-8x, Hoya, Tokyo, Japan), then spatially filtered using a microscope objective [8 mm effective focal length (EFL)] and a 50 μm pinhole, and finally collimated using a 250 mm focal length achromat (Thorlabs, Newton, NJ). The collimated laser beam is projected into the eye via the adjustable fold mirror. The single-pass retinal images [point spread function (PSF)]), captured with the retinal camera (PL B777F, PixeLink) in the physical model eye, were processed with custom-written Matlab routines. For every acquisition, 10 image frames were averaged to reduce pixel noise. Each averaged image was translated to position the image brightness peak at its center, and then cropped to 256 × 256 pixels, producing a record of the point spread function (PSF). Two-dimensional Fourier transformation of the PSF yielded the optical transfer function (OTF).

The analytical form of contrast sensitivity function (CSF) used for all optical modeling purposes was proposed by Mannos and Sakrison30 and is written as CSF (f) = 2.6 (0.0192 + 0.114 f) e^ {−(0.114 f)^(1.1)}, where “f” denotes spatial frequency in cycles per degree (c/deg). The real part of the OTF weighted with the CSF was calculated and truncated to 30 c/deg spatial frequency. As described elsewhere,31 we calculated the visual Strehl ratio (VSOTF), defined as the ratio of the volumes under the product of OTF and CSF with and without the aberration. For the no-aberration condition, the physical model eye set to emmetropia and having a 2 mm pupil was used to obtain the OTF. The closer the VSOTF measure is to 1.0, the better is the optical performance of the lens.

The optical performances of the CL and physical model eye combinations were measured in terms of VSOTF at three pupil diameters (3, 4, and 5 mm). In addition to the optical performance of correcting CLs centered on the physical model eye, performance at four decentered positions along the principle meridians was also evaluated. The coordinates of the CL center for the four positions, relative to the axis of the physical model eyes, were −0.50 mm, 0.00 mm; −0.25 mm, 0.00 mm; 0.00 mm, 0.25 mm; and 0.00 mm, 0.50 mm, as determined by the en face camera.

Optical performance was evaluated by varying object vergence, which is introduced by interchanging spherical trial lenses in front of the model eye. Unfortunately, the construction of the model eye did not provide enough space to add trial lenses close to the corneal plane. As an alternative, the trial lenses were placed next to the collimating doublet on the optic bench. An additional lens holder fixed to the mounting ring of the achromat held these trial lenses in place. Given that the trial lens plane is approximately 15 cm from the cornea, the required power of the lenses was calculated and selected to attain the required amounts of target vergence at the corneal plane.

MFCLs with high-add power (HA) and low-add power (LA) were tested. The HA designs and single-vision control CLs were tested at six target vergences, ranging from 0.00 D to −2.50 D in 0.50 D steps, whereas LA designs were tested at four vergences, ranging from 0.00 D to −1.50 D in 0.50 D steps. For each combination of lens design and target vergence, VSOTF was calculated. The set of VSOTF data across the range of target vergence was used to interpolate performance along the range 0.00 D to 2.50 D (1.50 D, if low add), using a cubic spline method to yield a function relating VSOTF with target vergence. The area under this function, termed AUTFVSOFT (area under through-focus visual Strehl ration), in units of diopters, was obtained using the trapezoidal rule of integration.

The custom-written software for the physical model eye enabled the user to adjust basic response attributes of the “retina” sensor, such as gamma, gain, and exposure time. A gamma value of 2.20 and gain factor of 0.00 were used throughout the investigation. Exposure time was set manually to avoid saturation of the sensor. Two factors determined optimum exposure time: pupil diameter and target vergence. A series of pilot studies were carried out to identify the relationship between exposure time and pupil size and target vergence that ensures fewer than five pixels exist with peak intensity values for any PSFs recorded. The pilot tests demonstrated that exposure time and pupil sizes were linearly related, with the slopes of the linear equations being dependent on target vergence. When the target vergences were between (i) 0 and −1 D, (ii) −1 D and 2 D, and (iii) −2 D and −3 D, the appropriate exposure time (milliseconds) of the retinal sensor that produced acceptable PSFs were given by 0.575 to 0.075*(pupil size), 0.950 to 0.150*(pupil size), and 1.350 to 0.200*(pupil size), respectively. The relationship to calculate optimum exposure time was integrated into the control software.

Back to Top | Article Outline

Effect of Lens Hydration and Role of Humidity Chamber on the Optical Performance

Dehydration of soft CLs can quickly lead to distortions in shape and degradation of the optical image quality. The developed physical model eye has neither tear fluid nor an eyelid to spread the tears. Therefore, soft CLs placed on the model eye were susceptible to rapid dehydration. To retard fluid evaporation from soft lenses, two methods have been implemented. First, as the model eye lacks tear fluid, saline (0.9% NaCl solution) was used as an artificial tear fluid. Second, the instrument has been equipped with a special humidity chamber (HC) over the cornea.

After placing the correcting soft CL on the cornea, regular instillation of saline was used to keep the lenses moist. To do this, saline from a 5 ml piston pump equipped with a soft-edge needle is released very slowly into the front portion of the model eye, which forms a reservoir. This process slowly submerses the sclera, cornea, and correcting CL under the artificial tear fluid. The reservoir is then emptied by slowly retracting the piston pump, taking care not to disturb the position of the CL. This procedure will be referred to as an effective blink action (EBA).

To test the efficacy of the HC, two soft single-vision CLs, Focus Dailies and PureVision, with a prescription of −2.00 D were used to correct the model eye set to −2.00 D myopia. The single-pass PSF method was used for this pilot assessment. After placing the correcting CL on the cornea, an EBA was performed. Thereafter, 50 s of videos of the PSFs, at the rate of 1 frame/s, were captured. The information obtained from the first 5 s, which represents the period for stabilization of the artificial tear film after the EBA, was not included in the analysis. The remaining 45 s PSF videos were Fourier transformed to yield their respective MTFs (modulation transfer function), and further, their volumes (area under MTF by time) were calculated using a trapezoidal rule of integration. The optical performance of each lens was gauged in terms of a time-dependent change of the peak-normalized volume of MTF, at 3 mm pupil, measured with and without using the HC. The average lapsed time before an optically significant change, defined as ≥10% drop in the peak-normalized MTF, was approximately 8 to 12 s, increasing to approximately 24 to 30 s using the HC (excluding the first 5 seconds of lens stabilization). In addition to retarding dehydration, the integrated heating unit in the HC prevented fogging of the glass window, providing unimpeded optical measurements. Overall, the HC was successful in maintaining hydration in the CLs for both the lenses used. One observation from this pilot assessment is that the evaporation rate of new soft CL taken straight out of its blister pack was found to be lower than that of reused lenses. A thorough cleaning of reused lenses, with a regular multipurpose solution, followed with fresh saline, produced evaporation rates similar to new lenses. Thus, for all optical performance measurements conducted in this work, thorough cleaning of lenses was adhered to as a protocol. A digital stop clock marked the lapse of 20 s during the investigation of optical performance using the various measurements. Once the 20 s time frame has lapsed, an EBA was performed before continuing with the performance evaluation. After an EBA, a wait period of 3 to 5 s was included to allow stabilization of the artificial tear film over the correcting CL.

Back to Top | Article Outline

Efficiency of the Centration and Decentration Tools

The en face camera was used to confirm centration or determine deliberate decentration of the correcting CLs. However, attempts to detect automatically the edge of the CL failed owing to three reasons: (a) low contrast of the CL edge against the model eye sclera background, (b) incomplete anodization with patchy areas on the sclera prevented continuous distinct appearance of the CL edge, and (c) the lathe marks on the sclera masked the edge outline. Therefore, a manual method had to be implemented for determining the position of the CL relative to the cornea. In this manual method, a still image from the live video stream of the anterior view is acquired. This captured image is automatically processed using standard filtering tools to enhance the edge contrast. At least four points on the corneal edge are manually selected. A circle is fitted through the selected points using Taubin method.32 The center of the circle is recorded as the optical axis, and its diameter is used to determine the “pixel to mm” scale factor of the image. The CL edge is identified by manually selecting at least six points and fitting a circle through them. The distance and direction between the corneal and CL centers are calculated and reported. Overall, the visibility of the CL edge was found to be dependent on the CL material, silicone hydrogel CLs edge traces being more readily than the others. For non-silicone hydrogel CLs, performing an EBA just before imaging facilitated edge detection.

Back to Top | Article Outline

Statistical Tests

The data obtained in terms of optical performance metrics were analyzed using univariate analysis of variance, with VSOTF as the dependent variable and pupil, target vergence, and lens type entered as fixed factors, for both HA and LA data sets. Type I error rate was set at p = 0.05. A statistically significant difference in VSOTF was observed between different pupils, accommodative states, and lens designs. The three-way interaction between pupils, target vergence, and lens types were also significant. Subsequently, the data were separated according to target vergence and pupil and reanalyzed, with VSOTF as dependent variable and lens type as a fixed factor. The differences between lens types at each pupil size and target vergences were tested as post hoc analyses using Bonferroni correction. Univariate analyses along the same lines were also performed with HA/LA data separated according to pupil and lens type, considering AUTVSOTF as the dependent variable and decentration as the fixed factor.

Back to Top | Article Outline


Optical performance for all test lenses was evaluated using two metrics: (i) VSOTF as a function of target vergence and (ii) AUTFVSOTF as a function of lens centration. The results will be presented under three sections. Each section is divided according to the add power of the correcting MFCL, either HA or LA, and three pupil sizes. Throughout these sections, the eight designs are coded as follows: the aspheric multifocals PureVision (white bars), AirOptix (dark gray bars), Focus Progressives (light gray bars), Proclear Distant (bars with vertical lines), Proclear Near (bars with horizontal lines); the concentric bifocals Acuvue (bars with dots), Acuvue Oasys (bars with checks); and lastly, the Focus Dailies single-vision CL (black bars). The height of the bars represents the mean, whereas the error bars indicate standard deviation (n = 4). For brevity, only statistically significant results will be discussed in the sections “Optical Performance of Well-centered Lenses” and “Optical Performance of Decentered Lenses.”

Back to Top | Article Outline

Optical Performance of Well-Centered Lenses

The figures pertaining to each subsection present the optical performances of select commercially available soft CLs in terms of VSOTF as a function of target vergences at three pupil diameters (3, 4, and 5 mm). The results are presented as a vertical montage of three graphs. Each graph represents the optical performance at a specific pupil size. The single-vision CL demonstrated the best VSOTF for distance at all pupils (Fig. 2a to c), being significantly different from all other test lenses at 3 and 4 mm pupils, and all lenses except Oasys bifocal at 5 mm pupil. VSOTF for distance were found to decrease linearly from 0.90 ± 0.03 to 0.20 ± 0.05 as pupil size increased from 3 to 5 mm. VSOTF for single-vision CL also decreased with increasing target vergence, with the rate of decrease being highest for 3 mm pupil.

Back to Top | Article Outline

HA Designs

Fig. 2 presents results for VSOTF for HA designs as a function of target vergence and pupil diameter. At 3 and 4 mm pupils (Fig. 2a, b), Proclear (D) design demonstrated good performance for intermediate test distance, although only significantly different from Acuvue bifocal and PureVision. At 4 and 5 mm pupils (Fig. 2b, c), Proclear (N) outperformed the rest for target vergences ≥−2.00 D. Overall, the relative differences between MFCLs decreased as pupil size increased to 5 mm. For far and intermediate test vergences, none of the MFCLs produced VSOTF that were significantly different from each other except for Oasys, for which the VSOTF was significantly greater than that for Proclear (N) and Acuvue designs at distance. For target vergences ≥−1.50 D (Fig. 2c), most tested MFCLs produced VSOTF that were approximately twice that obtained with the single-vision CL, indicating some incentive for near.

Back to Top | Article Outline

LA Designs

It can be observed from Fig. 3a that at a 3 mm pupil, the Proclear (D) design produced the best VSOTF for all target vergences except −1.50 D. Its performance was significantly better than other MFCLs at distance, but significantly better than only PureVision, AirOptix, and Focus progressives for target vergence of −0.50 D. At −1.50 D test vergence, the Proclear (N) design performed relatively better, although significantly better than only Oasys. The Oasys bifocal performed relatively better for distance, although significantly lower than the Proclear (D) design. At 4 mm pupil (Fig. 3b), the AirOptix design's performance was significantly greater than PureVision and Proclear (D) and (N) designs, for distance. Its performance became comparable with single vision for vergences >−0.50 D. The PureVision design produced the lowest performance measurements for distance but statistically significantly outperformed the rest for intermediate. Just like the HA designs, the relative differences between various LA MFCLs tested were found to be low in magnitude with an increase in pupil diameter to 5 mm.

Back to Top | Article Outline

Optical Performance of Decentered Lenses

On the living eye, soft CLs slide across the cornea; therefore, it would be of interest to analyze the performance of decentered lenses. Figs. 4 and 5 present the AUTFVSOTF of test lenses as a function of lens centration at three pupil diameters (3, 4, and 5 mm). Throughout this section, AUTFVSOTF was also loosely referred to as “through-focus” performance. The changes in AUTFVSOTF measures with horizontal and vertical decentrations are also discussed in the context of symmetry with respect to well-centered CL position. The performance of the control single-vision CL was found to be relatively stable about the well-centered position across both horizontal and vertical meridians, at all pupil diameters.

Back to Top | Article Outline

HA Designs

Fig. 4 plots AUTFVSOTF for HA designs as a function of lens centration at three pupil diameters. Both PureVision and AirOptix designs apparently produces a symmetric drop in AUTFVSOTF with decentration, at 3 and 4 mm pupils; however, only measurements of decentered AirOptix lenses were significantly different from its performance when well centered. The performances of both PureVision and AirOptix became asymmetric at 5 mm. The PureVision design demonstrated a statistically significant increase in AUTFVSOTF from its well-centered performance (approximately 0.40 D with 0.50 mm vertical decentration). This did not occur with the AirOptix design. The aspheric multifocals Proclear (D), Proclear (N), and Focus progressives and the concentric bifocals Acuvue and Acuvue Oasys all demonstrated asymmetry in AUTFVSOTF as decentration increased, at all pupil diameters. Although most of these lenses' performance was sensitive to decentration (p < 0.05) at 4 and 5 mm pupils, only Oasys bifocal's performance was significantly different from well-centered performance at 3 mm pupils (p < 0.05).

Back to Top | Article Outline

LA Designs

Fig. 5 depicts AUTFVSOTF for LA designs as a function of decentration at three pupil diameters. Differences in AUTFVSOTF were found between the vertical and horizontal decentration for the aspheric multifocals Proclear (D), Focus Progressives, AirOptix, and PureVision for equal amounts of decentration at all pupil diameters. At 3 mm pupil (Fig. 5a), the relative differences between the decentered and well-centered AUTFVSOTF measures for all test lenses, except Acuvue bifocal, were significantly different. Both Oasys and Proclear (N) designs demonstrated a symmetric drop in performance with decentration from their respective well-centered AUTFVSOTF measurements. At 4 and 5 mm pupils (Fig. 5b, c), lens decentration had no substantial effect on AUTFVSOTF for any test lens, except for Proclear (D) and (N) at 4 mm pupil and Oasys bifocal at the 5 mm pupil test case.

Back to Top | Article Outline


The aim of this work was to evaluate the optical performance of select commercially available simultaneous-vision MFCLs by using a physical model eye. Overall, most MFCLs exhibited reduced peak performance at the distance vision setting but achieved substantially better levels of optical performance along the range of distance to near target vergences, relative to the single-vision control. In agreement with previous findings,3335 performance was dependent on add power, design, pupil size, and centration of the correcting CL.

In the current work, we evaluated the optical performance of a wide selection of multifocal CL designs. In particular, we have included newer current-generation designs, the performance of which has not been published previously. By using an objective consistent platform—that is, the single-pass technique with the physical model eye—for evaluation, this study provides a reliable comparison of the CL designs.

An ideal objective optical metric of retinal image quality should have good correlation with subjective metrics, facilitating a valid objective prediction of visual performance. Thibos et al.31 have proposed approximately 30 optical metrics to be useful in predicting the actual visual performance, and identified 6 metrics that can account for at least 75% of the variability in visual acuity. Based on their findings, VSOTF (visual Strehl ratio in the frequency domain) was identified to be the best predictor of visual performance, and hence is used in this article as the metric of optical quality.

Before comparing key results of the current study with those in the literature, some possible limitations of the in vitro method should be considered. By using the EBA, we have effectively eliminated any artifacts that may have been introduced through variation of lens hydration. However, soft CL might suffer shape deformations when placed onto an artificial cornea causing unwanted optical effects. As all lenses were evaluated on the same model cornea, which was designed to be similar in shape to an average real cornea36,37 (anterior radius = 7.75 mm, Q = −0.25, diameter = 11.50 mm) bonded into lathed scleral rings of 12 mm radial curvature and 18 mm outside diameter29; the relative effect of any such deformation was judged by the observer and were found to be negligible. Obvious perturbations, such as air bubble under the CL, or folds on the CL, detected with the aid of the en face camera, were immediately corrected.

The optical performance of any simultaneous-vision MFCL is dependent on the proportional distribution of the area of pupil occupied by each of the power zones, providing corrections for distance, intermediate, and near vision.38 In general, the performance in terms of VSOTF as a function of object vergence declined linearly with increase in pupil size, except for the few cases discussed later in the text. For conventional concentric simultaneous-vision bifocals, the effects of pupil size on the overall performance have been predicted by Charman.38,39 Theoretically speaking, for a 4 mm pupil, the center-near design should favor near vision and the center-distant design should favor distance vision, provided the lens is well-centered and apportions 50% of pupillary area to each of the distant and near zones. The Proclear (D) and (N) designs appear to satisfy the aforementioned condition, although it is achieved with a relatively unconventional design—the Proclear (D) design has a spherical central zone (distance) of 2.3 mm diameter, surrounded by an aspheric annular ring of 5 mm outer diameter, and finally by a spherical annular zone (near) to its optic zone boundary of 8.5 mm diameter. In contrast, the Proclear (N) design has a spherical central zone (near) of 1.7 mm diameter, surrounded by an aspheric annular ring of 5 mm diameter, and finally by a spherical annular zone of 8.5 mm diameter. From the aforementioned data, the predictions by Charman38,39 appears to hold true for both LA and HA designs at the 3 mm pupil. Any increase in pupil size would disturb the balance between the performance for distance and near.

The PureVision, AirOptix, and Focus Progressive designs may be classified as continuous aspheric MFCLs. From theory, it can be anticipated that for such designs, the disturbance of the balance between distance and near performance in terms of reduced contrast is expected for all pupil sizes. The severity of the image degradation in terms of contrast loss would be directly proportional to the area of the pupil that is sufficiently wide to be influenced by the different varifocal powers. As expected, the overall performance in terms of VSOTF as a function of target vergence for PureVision, AirOptix, and Focus Progressive designs significantly deteriorated with increase in pupil size. The only exception to this finding was for LA designs at 3 and 4 mm pupils, where VSOTF at 4 mm pupil was relatively greater than the 3 mm test case.

The Acuvue bifocal achieves simultaneous vision by using five concentric zones of alternating distance and near powers with a distance zone at the center. The size and spacing of each zone is intended to optimize vision in varying lighting conditions, a feature the manufacturer promotes as “pupil intelligence.” One recently launched presbyopic product by the same manufacturer, the Acuvue Oasys bifocal, was also evaluated in this study. When compared with the other MFCLs, the performance of both the Acuvue bifocal and Oasys designs were relatively pupil independent, somewhat supporting the manufacturer's claims, and especially so for the HA designs. For LA designs, counterintuitively, the performance measurements at 4 mm pupil were relatively better than those obtained at 3 mm pupil. Although these designs are primarily bifocals, they also exhibited acceptable performance for other intermediate target vergences as well.

Decentration of the CL had some effect on AUTFVSOTF measures for some of the MFCLs tested, especially when compared against the single-vision lens. Differences were statistically significant for the LA versions of AirOptix, Proclear (D), and Focus progressives at 3 mm pupil. The change in AUTFVSOTF with lens centration was volatile and apparently unpredictable for HA multizone designs, that is, Acuvue bifocal and Oasys, at all pupils.

Martin and Roorda have considered the effect of lens decentration of the Acuvue multizone bifocal on subjects' simulated optical performance, using the area under MTF from 1 to 24 c/deg as the gauging metric, for a range of object vergences.35 They found that the overall differences between centered and decentered lenses averaged approximately 5% with decentration of 0.35 mm in the horizontal direction. For a particular accommodative state, the maximum difference observed was 11%. Interestingly, they found that for a few vergences, the performance showed an improvement, whereas for others, the performance was degraded, and more importantly, this trend was not systematic. Although the present study considered a different gauging metric (AUTFVSOTF) to what has been used by Martin and Roorda, the findings are interrelated, and the results of the two studies appear comparable. Asymmetry in the lens designs or prism or other manufacturing defects in the lenses could produce the asymmetry caused in the AUTFVSOTF performance as a function of lens centration.

A recent study by Legras et al.34 examined the visual performance of Proclear (D) and (N) designs along the range of distance to near target vergences. Their results showed that equal numbers of subjects had better performance with the distance design and with the near design, despite all subjects having similar refractive errors (low levels of myopic astigmatism). Although the aim of their study was not specifically to examine the role of inherent wavefront aberration (WA) profile on visual performance, they have attributed the observed differences in visual performance to the difference between the subjects' WA profiles, especially the level of vertical coma observed within the subjects. Recently, using computer-assisted ray tracing, we predicted that eyes having the same refractive prescriptions, but different levels of inherent WA profiles, can perform differently when fitted with identical MFCLs.28 Therefore, the optical performance of lenses tested on the physical model eye may only be a representative of the “real world” living eyes having aberration profiles comparable with the model eye.29 Thus, it may be expected that a specific MFCL design would be more successful with some wearers than others. In future, we hope to extend our studies to assess the role of WA profile on the performance using the developed physical model eye.

Throughout this article, the area under VSOTF along the range of distance to near target vergences has been used as one of the gauging metric to assess the performance of CLs. This metric was first used by Nagy and Yoon.40 A potential limitation is the summarizing nature of this metric. For example, two MFCL designs might produce VSOTFs of (a) 0.75, 0.50, and 0.25; and (b) 0.25, 0.50, and 0.75; for distance, intermediate, and near vergences, respectively. Although these hypothetical designs are different in terms of VSOTF, their AUTFVSOTF measures would not be different. For this reason, this metric has not been used to compare the performance of relatively similar lens designs; instead, it has mainly been used to gauge the change in the overall optical performance with centration of the CLs.

Back to Top | Article Outline

Limitations of the Study

The current physical model eye appears to promise a broad evaluation of the optical performance characteristics of various vision-correction devices. Nevertheless, depending on the application, it may be considered inferior to having personalized (i.e., with geometries and parameters customized according to a specific subject) physical model eyes.26,41 Moreover, it is necessary to bear in mind that the physical model eye only mimics the optical performance. Psychophysical effects that contributes to visual performance cannot be modeled and would have to be evaluated using human trials, for example, for assessing the neural transfer function.42 One other limitation of this work is that analyses were limited to monochromatic aberrations. This work should be extended to include the effects of chromatic aberrations, as they provide more information on the real-world performance. As described in a previous article,29 the optical performance assessment of a test lens in conjunction with the model eye has been achieved using two different single-pass techniques, one using visual acuity (LogMAR) charts presented on a visual display unit as test objects and another by quantitative analysis of the OTF using standard optical bench methods. For both these approaches, a standard setting, based on the manufacturer's recommendations, for the complementary metal sensor at the retinal plane was used. Although this setting is appropriate for testing with a visual display unit source, it may not be so for the optical bench setup. Nevertheless, given that settings were kept constant throughout, we believe that the relative results between lens types should remain comparable.

Back to Top | Article Outline


The optical performance of select commercially available simultaneous-vision MFCLs was evaluated with a physical model eye. All the MFCLs demonstrated reduced peak performance at distance viewing relative to a single-vision control. Certain MFCLs tested produced better performance response along the range of distance to near target vergences. Performance was dependent on add power, design, pupil size, and centration. Well-centered, HA, MFCL designs performed more poorly than their LA counterparts at all pupil sizes. Increasing pupil size was accompanied by a linear decrease in performance. Decentration substantially affected performance, although not always for the worse. The ability to directly measure single-pass images at the retinal plane of the physical model eye under identical conditions has proven to be helpful in evaluating the relative effectiveness of various simultaneous-vision multifocal CLs.

Ravi C. Bakaraju

University of New South Wales

Level 5, Rupert Myers NW, Gate 14, Barker St

Kensington, NSW 2033



Back to Top | Article Outline


This research was supported by postgraduate scholarships from the University of NSW, Sydney, Australia, and the Brien Holden Vision Institute, Sydney, Australia, to the first author (RCB). RCB was also a recipient of William. C. Ezell fellowship from the American Optometric Foundation and postgraduate research awards from the Cornea and Contact Lens Society of Australia. This work has been presented in part at the Association of Research in Vision and Ophthalmology meeting, 2011, Fort Lauderdale, FL. The physical model eye is the subject of a patent application (PCT/AU2009/000791).

Back to Top | Article Outline


1. Charman WN. Restoring accommodation: a dream or an approaching reality? Ophthalmic Physiol Opt 2005;25:1–6.
2. Bennett ES. Contact lens correction of presbyopia. Clin Exp Optom 2008;91:265–78.
3. Bennett ES, Jurkus JM. Presbyopic correction. In: Bennett ES, Weissmann BA, eds. Clinical Contact Lens Practice, 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005:531–48.
4. Evans BJ. Monovision: a review. Ophthalmic Physiol Opt 2007;27:417–39.
5. Freeman MH, Charman WN. An exploration of modified monovision with diffractive bifocal contact lenses. Cont Lens Anterior Eye 2007;30:189–96.
6. González-Méijome JM, Jorge J, Almeida JB, Parafita MA. Contact lens fitting profile in Portugal in 2005: strategies for first fits and refits. Eye Contact Lens 2007;33:81–8.
7. Hough A. Soft bifocal contact lenses: the limits of performance. Cont Lens Anterior Eye 2002;25:161–75.
8. Koffler BH. Management of presbyopia with soft contact lenses. Ophthalmologica 2002;216(suppl 1):34–51.
9. Morgan PB, Efron N. A decade of contact lens prescribing trends in the United Kingdom (1996–2005). Cont Lens Anterior Eye 2006;29:59–68.
10. Rajagopalan AS, Bennett ES, Lakshminarayanan V. Visual performance of subjects wearing presbyopic contact lenses. Optom Vis Sci 2006;83:611–5.
11. Richdale K, Mitchell GL, Zadnik K. Comparison of multifocal and monovision soft contact lens corrections in patients with low-astigmatic presbyopia. Optom Vis Sci 2006;83:266–73.
12. Woods CA, Jones DA, Jones LW, Morgan PB. A seven year survey of the contact lens prescribing habits of Canadian optometrists. Optom Vis Sci 2007;84:505–10.
13. Morgan PB, Efron N. Contact lens correction of presbyopia. Cont Lens Anterior Eye 2009;32:191–2.
14. Morgan PB, Efron N, Helland M, Itoi M, Jones D, Nichols JJ, van der Worp E, Woods CA. Demographics of international contact lens prescribing. Cont Lens Anterior Eye 2010;33:27–9.
15. Back A, Grant T, Hine N. Comparative visual performance of three presbyopic contact lens corrections. Optom Vis Sci 1992;69:474–80.
16. Back AP, Holden BA, Hine NA. Correction of presbyopia with contact lenses: comparative success rates with three systems. Optom Vis Sci 1989;66:518–25.
17. Back AP, Woods RL, Holden BA. The comparative visual performance of monovision and various concentric bifocals. J Br Contact Lens Assoc 1987;10(suppl):46–8.
18. Bullimore MA, Jacobs RJ. Subjective and objective assessment of soft bifocal contact lens performance. Optom Vis Sci 1993;70:469–75.
19. Collins MJ, Brown B, Bowman KJ. Contrast sensitivity with contact lens corrections for presbyopia. Ophthalmic Physiol Opt 1989;9:133–8.
20. Papas E, Young G, Hearn K. Monovision versus soft diffractive bifocal contact lenses. Int Contact Lens Clin 1990;17:181–7.
21. Papas EB, Decenzo-Verbeten T, Fonn D, Holden BA, Kollbaum PS, Situ P, Tan J, Woods C. Utility of short-term evaluation of presbyopic contact lens performance. Eye Contact Lens 2009;35:144–8.
22. Sheedy JE, Harris MG, Gan CM. Does the presbyopic visual system adapt to contact lenses? Optom Vis Sci 1993;70:482–6.
23. Woods RL, Saunders JE, Port MJ. Optical performance of decentered bifocal contact lenses. Optom Vis Sci 1993;70:171–84.
24. Woods RL, Port MJ, Saunders JE. Concentric-design rigid bifocal lenses, part II: visual performance. J Br Contact Lens Assoc 1993;16:71–82.
25. Bakaraju RC, Ehrmann K, Papas E, Ho A. Finite schematic eye models and their accuracy to in-vivo data. Vision Res 2008;48:1681–94.
26. Donnelly W 3rd. The Advanced Human Eye Model (AHEM): a personal binocular eye modeling system inclusive of refraction, diffraction, and scatter. J Refract Surg 2008;24:976–83.
27. Bakaraju RC, Ehrmann K, Ho A, Papas EB. Pantoscopic tilt in spectacle-corrected myopia and its effect on peripheral refraction. Ophthalmic Physiol Opt 2008;28:538–49.
28. Bakaraju RC, Ehrmann K, Ho A, Papas E. Inherent ocular spherical aberration and multifocal contact lens optical performance. Optom Vis Sci 2010;87:1009–22.
29. Bakaraju RC, Ehrmann K, Falk D, Ho A, Papas E. Physical human model eye and methods of its use to analyse optical performance of soft contact lenses. Opt Express 2010;18:16868–82.
30. Mannos JL, Sakrison DJ. The effects of visual fidelity criterion on the encoding of images. IEEE Trans Inf Theory 1974;20:525–36.
31. Thibos LN, Hong X, Bradley A, Applegate RA. Accuracy and precision of objective refraction from wavefront aberrations. J Vis 2004;4:329–51.
32. Taubin G. Estimation of planar curves, surfaces and nonplanar space curves defined by implicit equations with applications to edge and range image segmentation. IEEE Trans Pattern Anal Mach Intell 1991;13:1115–38.
33. Bradley A, Abdul Rahman H, Soni PS, Zhang X. Effects of target distance and pupil size on letter contrast sensitivity with simultaneous vision bifocal contact lenses. Optom Vis Sci 1993;70:476–81.
34. Legras R, Benard Y, Rouger H. Through-focus visual performance measurements and predictions with multifocal contact lenses. Vision Res 2010;50:1185–93.
35. Martin JA, Roorda A. Predicting and assessing visual performance with multizone bifocal contact lenses. Optom Vis Sci 2003;80:812–19.
36. Dubbelman M, Sicam VA, van der Heijde GL. The shape of the anterior and posterior surface of the aging human cornea. Vision Res 2006;46:993–1001.
37. Read SA, Collins MJ, Carney LG, Franklin RJ. The topography of the central and peripheral cornea. Invest Ophthalmol Vis Sci 2006;47:1404–15.
38. Charman WN. Theoretical aspects of concentric varifocal lenses. Ophthalmic Physiol Opt 1982;2:75–86.
39. Charman WN, Saunders B. Theoretical and practical factors influencing the optical performance of contact lenses for the presbyope. J Br Contact Lens Assoc 1990;13:67–75.
40. Nagy LJ, Yoon G. Interaction between primary and secondary spherical aberrations and its impact on retinal image quality through focus. Invest Ophthalmol Vis Sci 2008;49:E-Abstract 2422.
41. Navarro R, Gonzalez L, Hernandez-Matamoros JL. On the prediction of optical aberrations by personalized eye models. Optom Vis Sci 2006;83:371–81.
42. Dalimier E, Dainty C. Use of a customized vision model to analyze the effects of higher-order ocular aberrations and neural filtering on contrast threshold performance. J Opt Soc Am (A) 2008;25:2078–87.

presbyopia; multifocal contact lenses; physical model eye; optical performance

© 2012 American Academy of Optometry