Multifocal intraocular lenses (IOLs) have significantly advanced the field of cataract and refractive surgery. Until recently, multifocal IOLs were only been available in bifocal and trifocal designs, which provide multiple (two or three) distinct foci. Extended depth of focus (EDOF) IOL designs have recently emerged that create an elongated focal point to enhance the range of vision.1,2
The optical performance of the pseudophakic eye has been extensively studied. In addition to the benefits from the correction of spherical aberration, the reduction of chromatic aberration could further enhance the visual performance.3–6 Chromatic aberration is distinguished by longitudinal chromatic aberration (LCA) and transverse chromatic aberration (TCA).7,8 LCA characterizes the inability of a lens to focus different wavelengths at the same focal plane, whereas TCA describes wavelength-related changes to the image size of an off-axis object.7,8 In optical engineering, LCA is typically corrected with an achromatic doublet that consists of two cemented lenses having different dispersion.7 However, this approach could not be directly translated into IOLs due to technological limitations, so the use of diffractive optics appears to be the most suitable way to reduce LCA of IOLs.3,9,10 Given that refractive and diffractive lenses show opposite LCA behaviors,9,10 it is essential to understand how refractive and diffractive IOLs affect LCA, and thus the polychromatic image quality.
The aim of this study was to assess the effect of LCA of multifocal IOLs on polychromatic image quality and their potential to correct the LCA of the pseudophakic eye. To this end, we measured in vitro monochromatic and polychromatic modulation transfer functions (MTFs) of multifocal IOLs with different designs.
Materials and Methods
Table 1 shows the characteristics of the studied multifocal and monofocal IOLs. We included four multi-focal models with different optical designs: Mini Well Ready (SIFI MedTech, Sant'Antonio, Italy), AcrySof Restor SN6AD1 (Alcon Laboratories, Inc., Fort Worth, TX), AT Lisa 809MP (Carl Zeiss Meditec AG, Jena, Germany), and AT Lara 829MP (Carl Zeiss Meditec AG). The Mini Well is a refractive biconvex EDOF lens that uses spherical aberration to increase depth of focus. The Restor is a bifocal refractive-diffractive IOL with an apodized diffractive design that changes the energy split between the two foci with the pupil size. At a 3-mm aperture, the Restor allocates 70% of light to the far focus and 30% to the near focus. The AT Lisa is a bifocal full diffractive IOL that also shows asymmetric light distribution for far (65%) and near (35%). The AT Lara is an EDOF IOL that has a diffractive lens with an aspheric “aberration neutral” base platform and an optical design to correct chromatic aberration.
Characteristics of the Studied IOLs
LCAs of the four multifocal IOLs were compared with those of their monofocal counterparts by the same manufacturers. The Mini Well IOL was compared with the Mini 4 IOL (SIFI MedTech), the Restor IOL with the SN60WF IOL (Alcon Laboratories, Inc.), and the AT Lisa and AT Lara IOLs with the CT Asphina 409MP IOL (Carl Zeiss Meditec AG).
The optical performance of the IOLs was assessed using an OptiSpheric IOL PRO 2 optical bench (Tri-optics GmbH, Wedel, Germany). This device measures the nominal power and the MTF of IOLs with an accuracy of 0.1% to 0.3% and 2%, respectively. The IOLs were submerged in a balanced salt solution in a mechanical holder with two flat windows. A model cornea was a singlet lens with a positive spherical aberration of 0.28 μm. In this study, however, the IOLs were measured with a 3-mm aperture to minimize the effect of spherical aberration and pupil dependency of multifocal IOLs.11 A collimated beam of a LED source was used to illuminate two perpendicular fine slits (a cross reticle) that served as a test target. An image of the cross was projected by the model eye (with the IOL) onto a charge-coupled device camera (VA-1MCM120-A0-C; Vision Systems Technology, Vista, CA) through a microscope objective lens. As a result, two-line spread functions were obtained to evaluate sagittal and tangential MTFs. Given the rotational symmetry of the studied lenses, sagittal and tangential MTFs were averaged.
MTF results were presented graphically up to 100 lp/mm because this frequency corresponds approximately to a visual acuity of 20/20. The through-focus MTF was assessed in a defocus range from +2.00 up to −6.00 diopters (D). Moreover, the IOLs were compared by calculating the area under the MTF.12 The MTF area was analyzed at a range of spatial frequencies from 1 to 100 lp/mm (with 1 lp/mm sampling) using the following formula:
The MTF performance was assessed at the (best) far and near focus.
The MTF of the IOLs was measured in red, green, and blue light and in polychromatic light. To this end, we used three interference filters (10-nm bandwidth) with a central wavelength of 480, 546, and 644 nm and a photopic eye response filter that simulated the photopic luminosity function of the human eye. LCA of the IOL was calculated as the difference between the red and blue foci and expressed in diopters. In our set-up, LCA was 1.04 D for an “aphakic” model eye (ie, without the IOL). Each individual MTF and LCA measurement per condition was performed with one repetition. The standard deviation of the MTF assessment was tested for a discrete frequency of 50 lp/mm.
Polychromatic Image Simulation
The 1951 USAF resolution test chart was used to visualize the polychromatic image quality. Three separate photographs of the USAF target were taken with the three monochromatic filters at the position of an optimal polychromatic far and near focus. Given that the optical set-up featured a monochromatic camera, images were processed using a custom-made software (Image Processing Toolbox, MATLAB; Mathworks, Natick, MA) to add colors that corresponded to wavelengths of the monochromatic filters. These photographs were corrected for camera sensitivity and combined into one RGB (red, green, and blue) image using the same image processing software.
Table 2 presents the LCA values of the IOLs. The lens featured with chromatic aberration correction (AT Lara) demonstrated the lowest LCA. The Restor IOL showed higher chromatic dispersion than the other IOLs at far. All diffractive IOLs demonstrated reduced LCA levels at near compared to far focus. The Mini Well IOL showed a slightly higher LCA at near than at far.
Longitudinal Chromatic Aberration (Mean ± SD, D) of the Studied Multifocal IOLs and Their Monofocal Counterparts
The LCA level of the CT Asphina IOL was similar to that of the AT Lisa IOL but higher than that of the AT Lara IOL. LCA was found to be slightly lower in the SN60WF IOL than in the Restor IOL by 0.05 D. The Mini 4 IOL showed a higher LCA value than that of the Mini Well IOL.
The IOL Pro 2 devices showed good repeatability of the MTF assessment with a standard deviation of 0.001 or less for two consecutive measurements. Figure 1 presents in detail the MTF performance of the four multifocal IOLs. All three diffractive IOLs showed a spectral dependence of the light distribution for far and near demonstrating far dominance in red light and near dominance in blue light. The refractive lens showed only small differences in MTF results obtained at the three wavelengths.
Optical quality metrics of the multifocal intraocular lenses (IOLs). The figure presents three measure outcomes. The left panels show modulation transfer function (MTF) curves measured at the best far and near focus of each lens. The middle panels show MTF area values assessed at the far (black bars) and near (gray bars) focus. The right panels present the through-focus MTF evaluated at a single frequency of 50 lp/mm and for a +2.00 to −6.00 diopters (D) defocus range. The blue, green, red, and black lines correspond to MTF values measured with 480-, 546-, and 644-nm and polychromatic filters, respectively. The solid lines stand for the far focus MTF; the dashed lines stand for the near focus MTF.
In all lenses, the polychromatic MTF was worse than that measured in green light at the far focus (Figure 1). The percentage of the MTF area loss at the far focus was 14% for the AT Lara IOL, 27% for the AT Lisa IOL, 25% for the Mini Well IOL, and 34% for the Restor IOL. At the near focus, the AT Lara IOL demonstrated slightly better optical performance in polychromatic light by 5%. The MTF area value was lower in polychromatic light by 1% for the AT Lisa IOL, 14% for the Restor IOL, and 5% for the Mini Well IOL.
The through-focus scan presented in Figure 1 shows the position of monochromatic and polychromatic foci. All but one lens demonstrated a clear separation of the monochromatic (red, green, and blue) peaks at the zero defocus level. The AT Lara IOL showed nearly overlapping peaks at far and near, indicating a low LCA. The two other diffractive IOLs showed a smaller separation between their monochromatic near foci compared to that at far. For the Mini Well IOL, the separation of the monochromatic foci was distinct at far and near.
Polychromatic Image Simulation
The through-focus photographs of the USAF resolution chart are presented in Figure A (available in the online version of this article). Characteristic “fringes” of color appear in all simulated RGB images. The original polychromatic photographs and the RGB simulations show similar optical quality. Figure A confirms the MTF results of the AT Lara IOL showing a better image quality in red than in blue light at far, and the reverse relationship at near. At the far focus of the AT Lisa and Restor IOLs, the blue and red photographs appear blurred compared to the green ones. However, the image blur was reduced at the near focus, particularly for the AT Lisa IOL, indicating a lower LCA value. For the Mini Well IOL (Figure A), the blue and red photographs appeared to be out of focus at far, but it became less apparent at near.
1951 USAF resolution test chart photographs taken through the multifocal intraocular lenses. For more details on the image acquisition and processing, see the Methods section. The AT Lara and AT Lisa are manufactured by Carl Zeiss Meditec AG, Jena, Germany. The Restor is manufactured by Alcon Laboratories, Inc., Fort Worth, Texas. The Mini Well is manufactured by SIFI MedTech, Sant'Antonio, Italy.
We found that a diffractive optic IOL can be effective in correcting LCA of the pseudophakic eye. Moreover, we showed that uncorrected LCA may reduce the IOL optical quality in polychromatic light. Although the refractive IOL demonstrated comparable MTF levels in red, green, and blue light, the diffractive lenses showed a varying MTF performance depending on the wavelength.
The diffractive EDOF IOL demonstrated a clear potential for correcting IOL material dispersion and LCA of the eye. The model eye with the AT Lara IOL showed an LCA of 0.78 D at far, which was lower than that of the “aphakic” model eye (1.04 D), indicating the ability of the lens to compensate for chromatic aberration. This finding is in agreement with a study by Millán and Vega,10 who also showed an effective LCA correction of the Symfony IOL (Johnson & Johnson Vision, Jacksonville, FL). For the other lenses, without an intended chromatic aberration correction, LCA was consistently higher than that of the mechanical eye model due to a substantial contribution of the IOL material to LCA.
The highest LCA value was found in the Restor IOL (1.91 D). Although this lens showed a larger MTF area than the AT Lara IOL in green light, it was reversed in polychromatic light because the MTF area of the Restor IOL was 15% smaller than that of the AT Lara IOL (Figure 1). At the best near focus, this difference increased to 23%, as the AT Lara IOL showed an MTF improvement and a low LCA level (0.21 D). By contrast, the Restor IOL demonstrated a higher LCA of 1.05 D at near, and thus a worse MTF performance in polychromatic light. However, a larger difference might have been expected given that the dispersion level of the Restor IOL is more than twofold higher than that of the AT Lara IOL. The reason for this relatively small effect is the use of the photopic eye response filter, which simulates the spectral sensitivity of the human eye.13 In our experimental set-up, the photopic filter performed a spectral weighting that resulted in a lower intensity of wavelengths at the extreme ends of the spectrum (ie, 480 and 644 nm) compared to the 546-nm wavelength. As a consequence, the effect of LCA on the optical quality was diminished, yet the AT Lara IOL showed that the LCA correction can be of real benefit to the polychromatic image quality.
Although the LCA correction at the far focus emerges as a new feature of modern IOL designs, the compensation of the chromatic shift was found in all diffractive IOLs at the near focus, including those that were introduced more than a decade ago, such as the Restor IOL. For the diffractive apodized IOL, near LCA was lower than that at the far focus by 0.86 D. This LCA correction at the near focus results from the opposite sign of chromatic aberration in a diffractive and refractive element. LCA of a diffractive-refractive lens can be expressed as:
Although the LCArefractive component does not change at the far and near focus, LCAdiffractive varies between different diffraction orders (m). For example, the diffractive element of the Restor IOL directs the light energy to the far and near focus by using the zero (mFar = 0) and first (mNear = 1) diffraction orders, respectively.14 Given that the diffractive element has no power at the zero order, LCAdiffractive is zero. Therefore, in this case, LCA at the far focus depends only on chromatic aberration of a monofocal lens platform, which can explain close (far) LCA levels of the Restor IOL and the SN60WF IOL. At the near focus (mNear = 1), the Restor IOL has a 3.00-diopter power (P0) at the designed wavelength (λ0) of 550 nm.14 However, the power (P) of the diffractive element changes at different (than designed) wavelengths (λ)15 according to this formula:
For the wavelengths used in this study, P (λ = 480 nm) = 2.62 D and P (λ = 644 nm) = 3.51 D can be calculated, which results in an LCAdiffractive value of −0.89 D. So, for the measured LCArefractive = 1.91 D and the estimated LCAdiffractive = −0.89 D, LCA at the near focus would be 1.02 D, which is close to the measured value of 1.05 D. This approach can also be applied to predict the LCA level at the near focus of the AT Lisa IOL. Then the result would be 0.28 D compared to the value of 0.26 D found in the current study. Given that the AT Lara IOL provides LCA correction at the two foci, its optical design differs from a standard (mFar = 0, mNear = 1) diffraction order approach.10 In contrast to the other diffractive IOL included in this study, the diffractive element of this EDOF IOL seems to provide an add power to the two foci because LCA correction at the far focus can only take place for a non-zero LCAdiffractive component. A similar novel approach has also been introduced in the Symfony IOL (Johnson & Johnson Vision), as shown by Millán and Vega,10 indicating that a concept of the IOL correcting the eye's LCA is growing in popularity and becoming a new trend in the IOL market.
In contrast to its diffractive counterparts, the refractive IOL did not show a lower LCA value at the near focus. Although it yielded the highest LCA value at the near focus among all studied multifocal IOLs, the polychromatic MTF area was less reduced than that of the Restor IOL (5% vs 14%). The reason for that may be the EDOF character of the Mini Well IOL, which forms an extended near focus peak (Figure 1). Although the chromatic shift can be seen in the through-focus scan (Figure 1), the peak of each spectral component overlaps due to the EDOF effect, which appears to attenuate an effect of LCA of the IOL (Figure A). Intriguingly, the Mini Well IOL showed a different LCA at the far focus than its monofocal counterpart (Mini 4). This might suggest that the material dispersion of the Mini Well IOL differs from that of the Mini 4 IOL, but we could not confirm this explanation because the Abbe numbers of these IOLs have not been disclosed by the manufacturer.
The refractive multifocal IOL revealed close MTF results that were independent of the wavelength if the chromatic shift was accounted for (Figure 1). By contrast, the optical performance of the diffractive IOLs seems to strongly depend on the wavelength, as demonstrated in Figure 1. Although all diffractive optic IOLs showed a similar far focus dominance in green light, the MTF metrics differed in blue and red light. The AT Lara IOL demonstrated a strong distance vision dominance at 644 nm with a 3.8-fold larger MTF area at the far than at the near focus. However, a smaller but reverse effect was found at 480 nm with a 1.9-fold larger area under the MTF at near than at far. This spectral effect can also be noticed in Figure A (AT Lara). For the two other diffractive IOLs, the MTF metrics at the far and near focus were comparable in blue light and far dominant in red light. Given that the photographs of Figure A were taken at the best polychromatic focus, these changes to the monochromatic MTF performance of the AT Lisa and Restor IOLs were not clearly seen due to the chromatic shift, except from the near focus images of the AT Lisa IOL. The wavelength dependence could be explained by the diffraction efficiency at the mFar and mNear diffraction orders, which have been shown to be wavelength dependent.16 Portney16 proposed a geometric model to assess the light distribution of diffractive lenses in different wavelengths. We applied the proposed model to calculate the energy distribution of the AT Lisa IOL at the three wavelengths used in the current study. This resulted in a fraction of energy split of 0.63/0.37 for far/near at 546 nm, but that proportion changed to 0.51/0.49 and 0.74/0.26 at 480 and 644 nm, respectively. These values represent an ideal case without a light loss to other diffraction orders. Although the MTF quality metrics do not correspond in a one-to-one fashion with the light distribution, because the MTF can also be influenced by other factors, the MTF area of the AT Lisa IOL measured at the far/near focus (0.42/0.38 at 480 nm, 0.49/0.30 at 546 nm, and 0.57/0.22 at 644 nm) seems to show a similar behavior. This may indicate that the diffraction efficiency is an important factor affecting the optical performance of diffractive IOLs if they function in other than a designed wavelength. However, it remains to be elucidated whether spectral dependency noted has important functional effects on the patient's vision.
The analyzed monofocal and multifocal IOLs demonstrated a range of LCA levels that mostly depended on intrinsic properties of their biomaterial. Moreover, we showed that the diffractive IOLs can be effective in compensating for the dispersion of the IOL and the eye. Although considerable differences in LCA exist between the IOLs, the effect of chromatic aberration on the polychromatic image quality can be diminished by spectral weighting. The MTF performance of the diffractive IOLs showed a clear wavelength dependence, but a functional implication of this finding needs to be assessed in a clinical setting.