The advantage of extended-range lenses over other multifocal intraocular lens designs, in terms of reducing the occurrence of dysphotopsias or other disadvantages, has resulted in a fast proliferation of these types of lenses in the market. However, the American National Standards Institute (ANSI) has published clinical criteria for defining extended depth of focus (EDoF) IOLs [13]. It is worth noting that not all commercially available lenses, even if commonly referred to as EDoF, meet the ANSI criteria. However, the AcrySof IQ Vivity DFT015 model has been demonstrated to meet these criteria in clinical trials [14].

The primary feature of the lens under evaluation, in comparison to others available on the market, is an aspheric anterior surface with a central modification consisting in the addition of an optical element with a toroidal profile, resembling an axicon with a circular ring as described by McLeod [15], and is presented by the manufacturer as a non-diffractive wavefront shaping element (X-Wave technology). The optical fundamental of X-Wave technology is to engrave a toroid, with the rotation axis coaxial with the optical axis of the lens in one of the lens surfaces. The effect of this toroidal surface is to break the continuity of the wavefront. As a result of this discontinuous wavefront, the lens presents an extended depth-of-focus without the halos present when diffractive surfaces are employed.

To our knowledge, this is the first study to show the power profiles for the Vivity™ lens as well as the spherical aberration and coma across its entire surface, computed from the power maps measured by deflectometry.

The analysis of the average power profile obtained in this study (Fig. 1) reveals that the optical power in the periphery is lower and more stable than in the center, with an oscillating (or waving-like) pattern in the central zone up to 2.20 mm. The measured power profiles closely align with the manufacturer’s stated design: For all the three lenses studied, the power at the optical center is 1 D above the nominal power, and then oscillates, reaching this maximum power around 0.35 and 0.87 mm away from the optical center. Afterwards, the power peaked at 1 mm and then steadily decreased to the nominal power at, approximately, 1.25 mm away from the optical center. We hypothesize that these two zones could correspond to the beginning and end of the toroidal-shaped optical modification on the central anterior surface of the IOL. Furthermore, the power increase at the IOL center (extending 0.12 mm from the IOL center, within the central 0.24 mm optic zone) could correspond to the manufacturer’s intended change in central curvature to avoid the hyperopic part of the extended focal range, resulting in an additional power of approximately 1.25 D above the nominal power. Compared with other extended-range lens designs, such as the RayOne EMV lens [16] (Rayner Intraocular Lenses, Ltd, UK) or the ISOPure lens [17] (BVI-Physiol laboratory, Belgium), the design of the Vivity™ lens is entirely novel. This distinction arises from the fact that the aforementioned lenses show continuous power profiles that exhibit an increase in power higher than the nominal value, either in the peripheral or central zone, contingent upon whether the lens’ spherical aberration is positive or negative. This stands in contrast to the oscillating power design of the Vivity™ model. The addition achieved with the Vivity lens is between + 2.00 and + 2.50 D above the nominal value, depending on this nominal power, which could allow the patient not only to have spectacle independence in intermediate vision, but also in most daily activities, especially with electronic devices that can be used beyond 40 cm, as clinical studies have shown [4, 18]. Schmid et al. [19] in a study in which they estimated the extended range of focus for various lenses established that it was 1.7 D for 3 and 4.5 mm for Vivity™. This value is close to the addition found in our study for the lens of + 20.00 D, being the lens power evaluated by Schmid of + 22.00 D. This addition is higher than that reported for other EDoF lenses such as the Tecnis Eyhance [20] (Johnson & Johnson, CA, USA), ISOPure [17] or LuxSmart [19] (Bausch & Lomb GmbH, Germany). The 0.30 D difference in addition between our results (+ 2.00 D Add for a + 20.00 D IOL) and those by Schmidt et al. (1.70 D Add for a + 22.00 D IOL) could be due to almost two reasons. First, the different measurement methodology: We directly measure the power profile from deflectometry, whereas Schmidt et al. derive addition by the difference in peak location in the through-focus MTF for a 50 lp/mm spatial frequency. Second, Schmidt et al. used an in situ model eye according to ISO 11979, with NaCl (n = 1.337) which was heated to 35 °C, while our simulations were performed with a Cornea ISO2. For all these reasons, our results were not directly comparable to Schmidt’s, even though the difference is low (0.30 D).

In relation to primary spherical aberration, it has been observed that, for the three nominal powers analyzed (+ 10.00, + 20.00, and + 30.00 D), the values remain similar up to the central 3 mm. It is from this point onward that the + 30.00 D lens exhibits a slight difference, likely stemming from a peripheral design alteration to accommodate the higher power. The reported -0.20 µm value for the spherical aberration of this lens corresponds to an optical zone ranging from 4 to 4.55 mm, depending on the nominal power of the IOL. For the sixth Z(6:0) and eighth order Z(8:0) spherical aberrations, the results are consistent across all three powers. Positive values are observed in the central 2 mm for both aberrations, transitioning to negative values up to approximately the 3 mm zone before increasing back to positive values. These alterations in spherical aberration align with the regions where changes in the power profiles are observed, likely associated with the presence or absence of the optical element with a toroidal profile, to achieve the extended depth of focus. The aberrometric changes associated with the position of the central ring render the aberrometric design more complex than in other designs, where SA changes tend to be more continuous, lacking abrupt variations [17, 21, 22].

Regarding the primary [Z(3:1)] and [Z(3:-1)] and secondary [Z(5:1)] and [Z(5:-1)] coma, their influence is minimal, as the positive and negative RMS peaks align with the area where the ring is situated. In any case, these maximum values are approximately 0.025 microns or even smaller. This limited impact of these aberrations becomes evident when analyzing higher order aberrations. For both 3 and 4.5 mm, the HOAs values closely resemble the spherical aberration values for various orders, indicating that spherical aberration is the predominant factor with significant weight in the aberration analysis. Baur et al. [23] assessed higher-order aberrations in various lenses and observed that the Vivity™ lens exhibited a symmetric distribution of HOAs from the center of the lens. Furthermore, Schmid et al. [21] determined that SA was the sole significant Zernike aberration in this lens design, in agreement with our results. However, they did not investigate the changes in SA as a function of the optical zone, as conducted in the present study.

Aberrations reported in our work by means of the Zernike coefficient values for the AcrySof IQ Vivity DFT015 IOL, shown in Figs. 2–3, represent values for SA and coma at each specific distance from the lens center. Reporting single values of SA, for instance, provides limited information regarding the differences in focusing between central and peripheral light rays. For example, a value of “-0.20 microns” for an optical zone of 4 mm does not provide confident information about the behavior of light passing through the central 3 mm. This is the reason why it is so important to report aberrometric profiles and not isolated values.

Al-Amri et al. evaluated the aberrations in vivo, after implanting the lens under study [24]. The mean value of the lens implanted in their study was 21.53 ± 2.27 D, so we can compare them with those obtained with the + 20.00 D lens. The RMS HOA values obtained in vivo are slightly higher than those obtained in our study for a 3 mm pupil (0.18 vs. 0.09). This difference may stem from the real eye potentially presenting a greater number of aberrations compared to those of the ISO2 cornea, in which HOA is determined solely by spherical aberration. Furthermore, the other aberrations are not directly comparable due to differences in pupil sizes and because the aberrations are not independently separated in the clinical study.

The metrics for optical quality primarily rely on the MTF. This function determines the contrast transmitted through the model eye containing an IOL in relation to spatial frequency and pupil size. In our simulations, both the MTF value and other visual quality parameters were derived from simulations that incorporated Zernike values measured with the NIMO device and were supplemented with the values of an ISO2 cornea in the IOL plane. The initial observation in our results is that, for the three powers analyzed, the MTF values exhibit a notable similarity when comparing the same aperture. Taking the 20.00 D lenses as the reference, the Vivity™ lens yields MTF values of 0.4 at 50 lp/mm, slightly better than the values reported by Schmid et al. [25] and closely resembling those obtained by Azor et al. [26] or Baur et al. [23] for a 3 mm optical zone. For comparison purposes, a standard monofocal IOL (Tecnis ZCB00; Johnson & Johnson Surgical Vision, Inc.) shows an MTF value for 50 lp/mm around 0.5 considering a pupil of 3.0 mm [23]. As the aperture diameter increases to 4.5 mm, as expected, there is a decline in the MTF values, with results hovering around 0.2—very similar to those reported by Borkenstein et al. [27] and Schmid et al. [25]. Although our study is based on simulations using aberrometric values acquired with the NIMO TR1504, the results are entirely comparable to those obtained through other methods involving direct measurements on an optical bench. For instance, Azor’s [26] optical bench includes a model comprising an artificial cornea with a SA of + 0.27 μm for a 6 mm diameter, an iris diaphragm, and a wet cell containing saline in which the intraocular lens is immersed. Schmid [25] used an imaging test bench with a direct imaging setup using an in situ eye model with NaCl (n = 1337) at 35 °C to simulate a human eye was employed.

The PSF simulation computed from our NIMO results is very similar to that reported by Baur et al. [28], who claim a light pattern distribution showing minimally increased light spread compared to a monofocal IOL (the Alcon SN60WF in their study). The same result is presented by Kohnen et al. [29].

Regarding wavefront error variations, we have obtained an angular oscillating pattern of the wavefront map for a 3 mm pupil very similar to the wavefront mapping presented by Schmid and Borkenstein [21] and Baur et al. [23].

Since this is an optical bench evaluation of the AcrySof IQ Vivity DFT015 IOL, even though we can claim that this lens extends the focal range, as evidenced by the power profile presented in Fig. 1 with power increments above the nominal power around + 2.00/ + 2.50 D, our results do not provide explicit evidence about halo perception once implanted. Nonetheless, we present PSF results computed from wavefront aberrations for IOLs with + 10.00 D, + 20.00 D, and + 30.00 D for pupil sizes of 3.0 mm and 4.5 mm (see Fig. 5). The PSF for all these combinations shows a concentrated pattern of light distribution with minimal surrounding spread, suggesting a possible low impact of halo perception. This result agrees with the PSF reported for a + 20.00 D Vivity IOL by Baur et al. [28], who found a similar pattern of light distribution in the PSF of the Vivity IOL compared to a monofocal one (Acrysof SN60WF, Alcon, Fort Worth, TX, USA), both measured in an eye model with an optical bench. Regarding clinical results, the multicountry study by Bala et al. [4] found a similar incidence of perceived halos between patients implanted with the AcrySof IQ Vivity DFT015 IOL and a monofocal one (Acrysof SN60WF, Alcon, Fort Worth, TX, USA), using a quality of vision questionnaire. Additionally, the work by Kohnen et al. [29] also studied the impact of halo in patients implanted with AcrySof IQ Vivity DFT015 IOL, using a high dynamic range halo measurement system, finding similar results for the EDoF IOL and the monofocal one (Acrysof IQ, Alcon, Fort Worth, TX, USA).

Finally, regarding the possible limitations of our work, it must be considered that all the results shown in this study were obtained with the lens perfectly aligned with the optical bench axis. Therefore, further investigations are required to examine the potential impact of decentration on the optical behavior of this lens. Both the power and aberrometric profiles presented herein could change with tilting and/or decentering, potentially resulting in poorer optical outcomes. Future optical bench studies should address this possibility. In addition to optical bench studies, clinical studies involving implanted patients should analyze the impact of lens centration and tilting in visual quality.

With respect to the NIMO device used in this work to measure the optical properties of the evaluated IOL, some authors have pointed possible limitations of obtaining power maps from measuring the fringe pattern distortion using phase-shifting techniques, especially in the optical center when measuring contact lenses, and using filter options in the NIMO software. Thus, Kim et al. have reported lower repeatability measurements in the central 0.5 mm chord for bifocal and multifocal contact lenses using a specific filter configuration [30]. Nonetheless, we have previously reported very good results using the NIMO optical bench for IOL characterization, with no filter option enabled [12].

On the other hand, the main limitations associated to the used of Fourier Optics for computing the image-quality parameters are as follows: (1) The program assumes the validity of the scalar diffraction theory, (2) the PSF is computed for the far-field, and (3) the custom limitations in terms of noise and resolution of the Fast Fourier Transform (FFT) algorithm. The first assumption is valid in our case as we do not consider polarization or other vector effects. Regarding the second one, we have checked that the value of the Fresnel number is high enough to guarantee the validity of the far-field approximation [31]. Finally, we have worked with the highest possible sampling to minimize the shortcomings of the FFT algorithm.