Introduction
Multifocal intraocular lens (IOL) design is continuously evolving in order to improve visual quality, achieve spectacle independence and increase patient satisfaction after cataract surgery [
1,
2]. Typically, these IOLs have been developed using different optical principles; there are currently two main types of multifocal optics in IOLs: diffractive and refractive. Although good binocular vision is often achieved in patients following multifocal IOL implantations, some limitations have also been described for different models [
3]. Photic phenomena are among the main complaints of patients with multifocal IOLs. Due to light scattering at diffractive steps, multifocal IOLs featuring diffractive optics appear to be more prone to inducing glare symptoms than refractive-multifocal IOLs [
4]. On the other hand, near vision tends to depend more on pupil size when using refractive IOLs with a concentrically allocated near-focus zone [
5]. Furthermore, the multizonal IOL design may lead to a loss of contrast sensitivity associated with the distribution of the total available light between several focal points [
3]. While clinical studies have found better contrast sensitivity after the implantation of diffractive IOLs rather than refractive ones, some authors have disagreed with these findings and reached the opposite conclusions [
6,
7]. The limitations of concentric-refractive technology may be addressed by the introduction of a multi-segmented approach with alternating focal zones. Laboratory research suggests that using segments instead of concentric rings ought to provide better optical quality [
8] with minimal photic phenomena observed clinically [
9].
In vivo studies have reported that diffractive-refractive trifocal IOLs expand the postoperative visual range, resulting in a high rate of spectacle freedom and patient satisfaction [
10,
11]. Although the multizonal-refractive lens may also be a viable alternative to a trifocal approach, this conjecture has yet to be verified in both clinical and laboratory settings.
Benchtop measurements are often performed by evaluating image quality from the light distribution and efficiency of various foci [
12]. The International Organization for Standardization (ISO-11979-2) recommends using monochromatic green light for the optical testing of multifocal IOLs [
13]. However, this monochromatic light may be insufficient to extrapolate the visual performance since routine visual tasks are performed in light composed of various wavelengths. Therefore, polychromatic light testing appears more representative of real-life conditions and, as demonstrated recently, provides good agreement with clinical metrics [
14].
The purpose of this study was to comprehensively evaluate the performance of the multizonal-refractive IOL and compare it to a conventional trifocal-lens model using optical-quality metrics derived from benchtop measurements in polychromatic light. The secondary objective was to quantify the impact of chromatic aberration on the IOLs’ imaging quality.
Methods
Intraocular Lenses
The following IOL models were investigated in this study: 570 Precizon Presbyopic NVA—a multizonal-refractive IOL, and AcrySof IQ PanOptix—a hybrid-diffractive trifocal IOL. Two lenses of each model were available for testing, each with the same refractive power of 20.0 diopters (D).
The 570 Precizon Presbyopic NVA (OPHTEC BV, Groningen, The Netherlands) is manufactured from hydroxyethyl methacrylate (HEMA) copolymerized with ethoxyethyl methacrylate (EOEMA). This hybrid hydrophobic-hydrophilic material has a refractive index of 1.46 and Abbe number of 47. It features an aspheric surface to compensate for positive spherical aberration (SA) of the cornea by 0.11 µm. The lens has a refractive design consisting of alternating optical zones, which converge the incident light into two principal foci, and 'continuous transitional focus' for intermediate vision. The Precizon's most prominent (central) zone is dedicated to far vision, and the near power is set at + 2.75 D (IOL plane). The lens deliberately lacks rotational symmetry as it has an uneven distribution of 11 refractive zones along the horizontal and vertical meridians.
The PanOptix (Alcon Inc., Fort Worth, TX, USA) is a refractive-diffractive trifocal lens made of hydrophobic AcrySof material with a refractive index of 1.55 and an Abbe number of 37. This lens has a 4.5-mm (non-apodized) diffractive structure in the lens center and a 1.5-mm refractive zone located at the lens periphery. The PanOptix converges half of the light to far and a quarter to each remaining focus. The lens has an intermediate and near ADD power of 2.17 D and 3.25 D (IOL plane), respectively, as well as an aspherical design yielding − 0.10 μm of SA.
Optical Setup and Metrics
The optical quality of the study IOLs was assessed using the OptiSpheric IOL PRO2 optical bench (Trioptics GmbH, Wedel, Germany), following the guidelines of the International Standard Organization using monochromatic (546 nm) light [
13]. In addition, the optical quality was assessed in polychromatic light to study the impact of longitudinal chromatic aberration (LCA) on the optical quality. A model cornea with a positive SA of 0.28 m at 5.15 mm and 1 D of LCA was used for the measurements [
15].
The LCA of the IOLs was obtained from the difference between the IOL's power measured in red (644 nm) and blue (480 nm) light without the corneal model and expressed in diopters according to the formula
$$\mathrm{LCA}={\mathrm{IOL power}}_{480}-{\mathrm{IOL power}}_{644} \left[\mathrm{D}\right].$$
The IOLs’ imaging performance was evaluated for an aperture size of 3 mm (photopic) and 4.5 mm (scotopic). The modulation transfer function (MTF) of each IOL placed in the model eye was measured at its best focus plane using a 50 lp/mm MTF criterion; the IOL's tolerance for defocus was then examined within the range of + 0.5 D to − 3.5 D at the spectacle plane. Since the multizonal lens lacks rotational symmetry, care was taken to ensure the reproducibility of the IOL position during testing. Figure S1 in the Supplementary Information visualizes the locations of the sagittal and tangential meridians and provides a schematic drawing of the IOL. The two meridians were, however, averaged for the analysis, given that all meridians, rather than a specific one, contribute to the retinal image formation. A comparison between the sagittal and tangential MTFs of the multizonal and diffractive IOL is presented in Fig. S2 in the Supplementary Information, which also confirms that there is good alignment of the meridians in the refractive-diffractive model.
Furthermore, the optical quality was also evaluated with the area under the MTF (MTFa), given its significance to visual quality [
16]. The MTFa, which was obtained by integrating the corresponding MTF values at a range of spatial frequencies from 1 to 50 lp/mm, was converted to simulated visual acuity (simVA) as described by Alarcon et al. [
16]. This article is based on bench evaluations and does not contain any studies with human participants or animals performed by any of the authors.
Discussion
This study demonstrates that, according to the MTF analysis, the multizonal-refractive approach does not fall short of the established trifocal IOL and can be used to extend the visual range of pseudophakic patients. Although the performance of both models was affected by LCA, the HEMA/EOEMA material had a lower LCA level than AcrySof. The diffractive model, however, showed a reduction in dispersion effects at its secondary foci.
The chromatic aberration of the phakic eye is caused by ocular-media light dispersion, with the cornea and the crystalline lens being the most significant contributors [
17]. Following phacoemulsification and IOL implantation, the amount of chromatic aberration is reduced compared to a phakic eye [
18], which shows variability between IOL models. One differentiating factor is the spectral dispersion of biomaterials, which show varying levels of wavelength dependency for their refractive indices [
19]. Another important factor is lens design, as the effects of chromatic aberration can be corrected through the diffractive principle [
15,
20]. Therefore, both the IOL material and the design may affect the pseudophakic eye's chromatic aberration, which, in turn, may compromise the polychromatic optical quality [
15,
20].
The impact of chromatic aberration for different types of IOLs has been extensively studied, both in vivo and in vitro [
18,
21‐
23]. Pérez-Merino et al. used a laser ray tracing aberrometer to measure the optical aberrations at wavelengths of 532 nm and 785 nm [
21]. They measured the chromatic difference of focus in nine eyes with the Tecnis (Abbe value 55) IOLs and nine eyes with the Acrysof (Abbe value 37) IOL and reported a significantly lower chromatic difference of focus value for Tecnis than for Acrysof (0.46 D and 0.75 D, respectively) [
21]. Siedlecki et al. measured the objective refraction at various wavelengths (470–660 nm) and used an adapted visual refractometer to evaluate the variation of the chromatic difference of refraction with the IOL material. That study reported that the properties of IOL biomaterials have a significant influence on the magnitude of the chromatic aberration of the pseudophakic eye [
23], which was also seen in the current study. Loicq et al. presented a bench study comparing the through-focus MTFs of nine lenses at different wavelengths of 480 nm, 546 nm, and 650 nm. They found that for non-zero diffractive orders, the refractive LCA contributions may be compensated by the reversed LCA of a diffractive element, with the efficiency dependent on the material and optical design [
22].
Our in vitro measurements of LCA for AcrySof IOL (0.91 ± 0.01 D) fall within the LCA range reported by Millan et al. using an autorefractor with a Scheiner disc illuminated by different wavelengths (0.96 ± 0.34 D) [
18]. Additionally, that study showed that the Tecnis IOL had a lower level of LCA than the AcrySof IOL, which is consistent with the IOL material's dispersive characteristics, since lower LCA values have been reported for materials featuring higher Abbe numbers. This conclusion also holds for our result, since the multizone HEMA/EOEMA IOL (Abbe value 47) had lower LCA values (0.41 D) than the diffractive AcrySof IOL [
18].
As a consequence of the LCA, the MTF values were significantly lower when measured under polychromatic illumination than those measured with monochromatic light. Our results show that the polychromatic MTF of the multizonal lens (24%) decreased less than that of the diffractive model (44%) at far. The material properties may be one factor, as the diffractive lens has a lower Abbe number and thus higher chromatic aberration than the multizonal lens [
24]. However, at the secondary foci, chromatic effects were minimized by the lens's diffractive design at the intermediate and near foci [
15,
20]. Therefore, as shown in Fig.
3, the LCA correction was more effective in the latter position. This finding agrees with the conclusions of Loicq et al. [
22], who also demonstrated a nearly-zero LCA contribution of the diffractive lens at the intermediate focus. In addition, the negative LCA at the near focus reported by Loicq et al. reduces the eye's LCA, which may explain the minimal differences between the monochromatic and polychromatic performance of the diffractive lens at about − 2.5 D. This correction can take place because refractive and diffractive optical elements yield LCAs of opposite signs, but a similar effect is not feasible in the multizonal lens due to its refractive design. Still, the multizonal-refractive lens’s performance was not substantially affected by LCA, which results from its lower material dispersion, specified by a high Abbe number, as confirmed by the results of the current study.
The PanOptix lens has been extensively studied, both on an optical bench and in clinical trials. Using the same benchtop setup and spectral conditions as in our study, Naujokaitis et al. compared the laboratory-derived defocus curves of the PanOptix with a complementary (EDoF) system [
25]. According to their findings, the trifocal lens had a simulated VA of 0.2 logMAR or better throughout the range of + 0.5 D to − 3.5 D, and they also identified three peaks at 0 D of focus (− 0.02 logMAR), − 1.75 D (0.03 logMAR), and − 2.5 D (0.02 logMAR) [
25]. Our results show equal or better VA for the same defocus range, but we observed lower values of simulated VA for far vision (0.00 logMAR) and near focus (0.06 logMAR). These minute discrepancies can be attributed to differences in the metrics used to derive the simulated VA. Naujokaitis et al. used a weighted optical transfer function, while the MTFa was used in our study as the parameter for VA prediction. Despite differences in the approaches used to calculate both metrics, they appear to show comparable results, and both can be used to simulate postoperative performance, as indicated by their high correlation with clinical VA [
16].
This agreement between predicted and clinical defocus curves can be found when comparing our data to those from the clinical literature [
26,
27]. For instance, a 6-month prospective case series study reported significant improvements in uncorrected and corrected VA outcomes at 1 month after the diffractive lens implantation. The monocular defocus curves showed that a VA equal to or better than 0.30 logMAR was maintained between + 0.50 D to − 3.00 D, with the best VA occurring at 0.00 D and − 2.50 D, which corresponds well with the results of our study [
27]. However, we also observed variations in reported defocus curves between clinical studies of the same model. Kohnen et al. reported the secondary VA peak of a diffractive lens at − 2.00 D in both monocular (0.01 logMAR) and binocular (− 0.02 logMAR) defocus curves, yielding weaker agreement with our in vitro results [
26]. In a clinical setting, however, non-IOL-related factors, such as intersubjective variability or measurement error, may affect defocus-curve results, leading to disagreement when laboratory findings collected under strictly controlled conditions are compared against clinical ones.
Eom et al. enrolled 40 patients implanted with the Precizon and assessed IOL-induced astigmatism [
28]. That study found an increase in astigmatism related to the IOL ranging from 0.68 ± 0.58 D to 1.05 ± 0.81 D, depending on the lens orientation. However, the authors concluded that it was not clinically significant given its low impact on manifest refraction. Still, this finding may support our observation of the astigmatic-like effect in the recorded USAF-target images. Secondary to the astigmatism assessment, Eom et al. reported a binocular defocus curve for their patients [
28]. Alió et al. also evaluated the defocus curve of ten (Precizon 570 NVA) patients [
29]. Despite differences in the intermediate VA level between the two studies, which may result from the small (10) study population analyzed by Alió's group, there are apparent similarities to the simulated VA in our study. Although zero-defocus VA was better than predicted from our optical measurements in those clinical studies, at − 1 D, the MTF-based prediction appears to overestimate VA. In our study, a VA value of 0.05 logMAR was predicted, but for Eom et al., it was 0.1 logMAR [
28]. We noted a similar difference in the prediction for PanOptix, which, besides clinical factors, may also indicate the need for refinement of the applied VA-prediction formula. Although the findings of this in vitro study cannot be extrapolated one-to-one into a clinical outlook, valuable insight into the optical properties of multizonal-refractive designs can be gained, and good agreement with the average effect observed clinically is obtained.
Acknowledgements
Donald J. Munro contributed to the review of the manuscript. Ethics committee approval was not required for this laboratory study.