The main purpose of cataract surgery is to provide the best quality of vision to patients. After the introduction of wavefront technology in ophthalmology, a primary area of interest has been the role of higher order aberrations and mainly spherical aberration as it relates to the quality of vision.
Classical intraocular lenses (IOLs) present spherical designs (with positive spherical aberration), whereas the latest designs incorporate aspheric designs (with negative spherical aberration or without spherical aberration) to compensate totally or partially for the positive spherical aberration in the cornea after cataract extraction. These aspheric designs are calculated for an averaged corneal profile of a population with normal corneas. However, it is important to take into account that millions of corneal refractive surgeries have been performed worldwide1 and after these procedures, corneal spherical aberration is modified. A hyperopic laser ablation changes the corneal asphericity towards more negative values.2–4 Therefore, it is of interest to know how current IOL models perform when combined with different amounts of negative spherical aberration produced by hyperopic ablations.
Adaptive optics technology represents a useful and non-invasive tool to properly assess the benefits of different IOL designs by measuring, correcting, and simulating aberration patterns.5–9 For this study, an adaptive optics visual simulator was used to assess the impact of change in corneal spherical aberration after hyperopic LASIK on the visual results obtained with different IOLs.
Patients and Methods
Ten individuals, aged between 21 and 30 years and experienced in psychophysical experiments, were included in the study. Spherical refractive errors ranged between −2.00 and +0.00 diopters (D) with astigmatism <0.50 D. They had no known ocular pathology. Approximately 30 minutes before experimental measurements, three drops of cyclopentolate hydrochloride 0.5% were instilled to paralyze accommodation.
The tenets of the Declaration of Helsinki were followed. Informed consent was obtained from each patient after verbal and written explanation of the nature and possible consequences of the study. The protocol received institutional review board approval.
For ensuring an optimum alignment of the pupil and apparatus, patient head movement was limited with a chin-and-forehead rest and a bite bar.
An adaptive optics visual simulator (crx1; Imagine Eyes, Orsay, France) was used for measuring and correcting ocular aberrations and inducing different aberration patterns. The residual level of aberrations of the system for static correction procedures was always set to root-mean-square (RMS) of higher order aberrations (up to 10th order) <0.03 μm. The adaptive optics system contains two basic components: the wavefront sensor (Shack-Hartmann) and wavefront corrector (deformable mirror) (Fig A, available as supplemental material in the PDF version of this article). The instrument permits the assessment of visual function through a microdisplay. The technical characteristics of the crx1 and the way in which it induces aberration patterns have been widely described in the scientific literature.5,7,8,10,11
Intraocular Lens Aberration Patterns
For this study, three IOLs with different designs were considered: the Akreos Adapt (Bausch & Lomb, Rochester, New York); AcrySof IQ SN60WF (Alcon Laboratories Inc, Ft Worth, Texas), and Triplato Y3601075 (AJL Ophthalmic, Álava, Spain). The Akreos Adapt and Triplato Y3601075 feature spherical designs, whereas the AcrySof IQ SN60WF has an aspheric posterior surface that is designed to reduce whole eye spherical aberration.
A commercial aberrometer (IRX-3, Imagine Eyes) together with a custom-made wet-cell was used to obtain the in vitro wavefront of the IOL (IOL+wet-cell). The aberrations of the wet-cell alone were also measured and subtracted from the IOL+wet-cell aberrations. Before taking the measurements of the IOLs, the repeatability and reproducibility of the system were studied performing 10 measurements of a monofocal IOL and an additional 10 measurements of the same IOL but extracting it from the wet-cell (artificial eye) and then introducing it again before and after the measurement. Modulation transfer function (MTF) variations were assessed and all changes were approximately 0.1 μm, which is considered highly satisfactory.12 At the same time, the accuracy of the system was assessed by three different methods obtaining results that confirm the accuracy of the system in terms of IOL aberration profile calculation (N. López-Gil et al, personal communication, 2008; N. López-Gil et al, personal communication, 2009; Bonaque et al, personal communication, 2009).
In vitro wavefronts of each IOL were obtained for five situations: centered, 0.2 mm and 0.4 mm of decentration, and 2° and 4° of tilt around the vertical axis. All measurements were taken for a 5.0-mm pupil size.
Corneal Aberration Patterns
Corneal higher order aberrations, which have been shown to be stable with age,13–15 were selected. Over this averaged profile, two different increments of spherical aberration induced by LASIK procedures for low and high hyperopia (−0.105 μm and −0.33 μm, for the low hyperopia group and high hyperopia group, respectively)2 were added. Corneal higher order aberrations from 3rd to 5th order were considered,15 being adjusted to a 5.0-mm pupil.16
The first step of the procedure was to capture the eye’s wavefront analysis using the crx1 after waiting 30 minutes from the instillation of three drops of cyclopentolate hydrochloride 0.5%. The dilated pupil diameter was checked, which was ⩾6 mm in all patients. The pupil size was optically set to 5.0 mm, using the simulator artificial pupil. When the eye’s wavefront up to 5th order was compensated, the adaptive optics system was used to apply the wavefront pattern of the three IOLs in the five situations previously described in both groups (Fig 1).
Figure 1. Zernike coefficients of the three intraocular lenses in the five situations studied (centered, 0.2 mm and 0.4 mm of decentration, and 2° and 4° of tilt) for the A) low hyperopia group and B) high hyperopia group at 5.0-mm pupil.
The visual acuity evaluation procedures were in accordance with other studies, and a Freiburg Acuity Test17 software package was used to measure the patient’s corrected monocular visual acuity at 10%, 50%, and 100% contrast in all situations through the adaptive optics system.5,18 The Freiburg Acuity Test with a white background luminance was adjusted to 50 cd/m2 and the software was run on a separate laptop connected to the adaptive optic system’s internal microdisplay monitor. A black Landolt C was presented to the patients and they indicated the orientation of the optotype on a numeric keypad.
Data were analyzed using SPSS for Windows v.17.0 (SPSS Inc, Chicago, Illinois). Normal distribution of variables was assessed using the Kolmogorov-Smirnov test. A repeated measures analysis of variance was used to gauge any statistically significant difference within the different situations. Post-hoc multiple comparison testing was performed using the Holm-Sidak method. Differences were considered statistically significant when P<.05 (ie, at the 5% level).
The study was performed in 10 eyes of 10 patients with a mean age of 25.25±3.77 years (range: 21 to 30 years). The outcomes found were divided in two groups, corresponding to the increase of spherical aberration provided by low and high hyperopic LASIK.
Low Hyperopia Group
Table 1 shows logMAR visual acuity values in all situations for the low hyperopia group. Values for the centered situation were similar between lenses being 20/20 or better for 100% contrast, and approximately 20/25 and 20/50 for 50% and 10% contrast, respectively. No statistically significant differences were found among the three IOLs at any contrast (P>.05).
Table 1: Visual Acuity at 100%, 50%, and 10% Contrast With Three Intraocular Lenses in the Five Situations Studied for the Low Hyperopia Group
Effect of Decentration. For the Akreos Adapt, visual acuity decreased when the IOL was decentered 0.2 mm only at 100% contrast (P=.006). A statistically significant decrease was found at all contrasts when the IOL was decentered 0.4 mm (P=.001, P=.002, and P=.029 at 100%, 50%, and 10% contrast, respectively). Statistically significant differences were found between 0.4- and 0.2-mm decentration (P=.008, P=.003, and P=.02 at 100%, 50%, and 10% contrast, respectively).
For the AcrySof IQ SN60WF, visual acuity decreased when the IOL was decentered 0.2 mm only at 100% contrast (P=.035). When the IOL was 0.4-mm decentered, visual acuity decreased at 100% and 50% contrast (P=.007 and P=.02, respectively). Differences were also found between 0.4- and 0.2-mm decentration at 100% and 50% contrast (P=.04 for both).
For the Triplato, visual acuity decreased at 100% and 50% contrast when the IOL was decentered 0.2 mm (P=.01 and P=.004, respectively). When the lens was decentered 0.4 mm, visual acuity decreased at all contrasts (P<.001 for every contrast). Between 0.4- and 0.2-mm decentration, statistically significant differences were found at 100% and 10% contrast (P<.001 and P=.02, respectively).
Figure 2A shows a comparison of visual acuity with each IOL studied for both degrees of decentration. For 0.2-mm decentration, the Akreos Adapt showed better results than the Triplato at 50% contrast (P=.01) and the AcrySof IQ SN60WF at 10% contrast (P=.04). For 0.4-mm decentration, all IOLs showed the worst outcomes at all situations. At 100% contrast, the Akreos Adapt (P=.006) and AcrySof IQ SN60WF (P=.02) obtained better results than the Triplato. At 50% contrast, no differences were found for the three IOLs. At 10% contrast, differences were found between the Triplato and AcrySof IQ 60SNWF (P=.02).
Figure 2. Visual acuity (VA logMAR) comparisons for the A) low hyperopia group and B) high hyperopia group at 100%, 50%, and 10% contrast with the Akreos Adapt, AcrySof SN60WF, and Triplato when the IOLs were centered, 0.2 mm and 0.4 mm decentered, and 2° and 4° tilted. Mean values are shown and error bars represent the standard deviation.
Effect of Tilt. For the Akreos Adapt, no statistically significant decrease of visual acuity was found at any contrast when the IOL was tilted 2°. When the IOL was tilted 4°, visual acuity decreased significantly only at 10% contrast (P=.03).
For the AcrySof IQ SN60WF, visual acuity decreased significantly only at 100% contrast when tilted 2° (P=.01). When the lens was tilted 4°, differences were found at all contrasts (P=.002, P<.001, and P=.04, at 100%, 50%, and 10% contrast, respectively).
For the Triplato, visual acuity decreased significantly when tilted 2° at 100% and 50% contrast (P=.006 and P<.001, respectively). Between 4° of tilt and centration, statistically significant differences were found at 100% and 50% contrast (P<.001 for both). Differences between 4° and 2° of tilt were only found at 100% contrast (P=.002).
Figure 2A also shows a comparison of visual acuity with each IOL for both degrees of tilt. For 2° of tilt, the Akreos Adapt obtained better results than the other two IOLs at 100% contrast (P=.01 for the AcrySof IQ SN60WF and P=.007 for the Triplato) and 50% contrast (P=.04 for the AcrySof IQ SN60WF and P=.03 for the Triplato). For 4° of tilt, the best results were also obtained with the Akreos Adapt (P=.005, P=.01, and P=.002 at 100%, 50%, and 10% contrast, respectively, for the AcrySof IQ SN60WF; and P<.001, P=.003, and P=.01 at 100%, 50%, and 10% contrast, respectively, for the Triplato).
High Hyperopia Group
Table 2 shows logMAR visual acuity in all situations for the high hyperopia group. The worst results were obtained with the Triplato at all contrasts (P=.006, P=.006, and P=.005 at 100%, 50%, and 10% contrast, respectively, for the Akreos Adapt; and P=.008, P=.02, and P=.04 at 100%, 50%, and 10% contrast, respectively, for the AcrySof IQ SN60WF).
Table 2: Visual Acuity at 100%, 50%, and 10% Contrast With Three Intraocular Lenses in the Five Situations Studied for the High Hyperopia Group
Effect of Decentration. For the Akreos Adapt, no differences were found between centration and 0.2-mm decentration at any contrast. When the lens was decentered 0.4 mm, differences were only found at 100% contrast (P=.01).
For the AcrySof IQ SN60WF, visual acuity decreased at all contrasts when the lens was decentered 0.2 mm (P=.01, P=.04, and P=.04 at 100%, 50%, and 10% contrast, respectively). When decentered 0.4 mm, differences were also found at all contrasts (P=.004 for 100%, P=.04 for 50%, and P=.002 for 10%). Between 0.4- and 0.2-mm decentration, differences were only found at 100% contrast (P=.04).
For the Triplato, visual acuity decreased significantly when the lens was decentered 0.2 mm only at 50% contrast (P=.03). Differences at 50% contrast were also found when the lens was decentered 0.4 mm (P=.03).
Figure 2B shows a comparison of visual acuity with each IOL studied for both degrees of decentration. For 0.2-mm decentration, the Akreos Adapt obtained better results than the other two IOLs at 100% and 10% contrast (P=.02 and P=.04 at 100% and 10%, respectively, for the AcrySof IQ SN60WF; and P=.04 and P=.034 at 100% and 10%, respectively, for the Triplato). For 0.4-mm decentration, visual acuity decreased for all IOLs at all contrasts. At 100% and 10% contrast, no differences were found among the three IOLs. At 50% contrast, differences were found between the Triplato and AcrySof IQ SN60WF (P=.04).
Effect of Tilt. For the Akreos Adapt, visual acuity decreased when tilted 2° at 100%, 50%, and 10% contrast (P=.004, P=.004, and P=.02, respectively). When the lens was tilted 4°, differences were found at 100% and 50% contrast (P<.001 and P=.003, respectively). Differences between 4° and 2° tilt were found only at 100% (P=.006).
For the AcrySof IQ SN60WF, visual acuity decreased when tilted 2° at 100% and 10% contrast (P=.003 and P=.03, respectively). When the lens was tilted 4°, differences were also found at 100% and 10% contrast (P=.02 for both).
For the Triplato, no differences were found when the IOLs were 2° or 4° tilted at any contrast. Between 4° and 2° of tilt, differences were only found at 100% contrast (P=.008).
Figure 2B also shows a comparison of visual acuity with each IOL studied for both degrees of tilt. For 2° of tilt, at 100% contrast, the AcrySof IQ SN60WF presented worse results than the Akreos Adapt (P=.04) and Triplato (P=.04). No differences among the three IOLs were found at 50% and 10% contrast. For 4° of tilt, no differences were found among the three IOLs at 100% and 50% contrast. At 10% contrast, differences were only found between the Akreos Adapt and AcrySof IQ SN60WF (P=.03).
The present study evaluated visual quality through adaptive optics visual simulation with three different IOL designs in patients with negative increments of corneal spherical aberration due to prior LASIK procedures for low and high hyperopia. One lens was aspheric, and the remaining two were spherical designs with different amounts of positive spherical aberration (see Fig 1).
Analyzing the results shown in Tables 1 and 2, it seems that the higher spherical aberration amount of change is related to the visual acuity reduction found in the high hyperopic group (ie, at 100% contrast, visual acuity for the Akreos Adapt, AcrySof IQ SN60WF, and Triplato decreased 0.09, 0.05, and 0.16 logMAR, respectively). For the low hyperopia group, visual acuity was comparable for the three IOLs. In this group, the range of residual spherical aberration was −0.037 to 0.109 μm (see Fig 1). Thus, taking these results into account, it could be suggested that this residual spherical aberration does not affect visual quality. This is in accordance with a previous study in which residual spherical aberration of ±0.1 μm did not decrease visual acuity significantly.18 For high hyperopic ablations, the residual spherical aberration becomes significantly more negative for all cases. In the high hyperopia group, the best results were obtained with the Akreos Adapt; these results could be explained due to the fact that this IOL has a higher positive spherical aberration value and compensates for the increase of negative spherical aberration after high hyperopic ablations.
Previous studies have shown that IOL design is not the only factor in achieving good visual quality after cataract surgery, with centration or tilt of the IOL having an important impact on visual performance.7,9,19–26 Therefore, in the present study, the adaptive optics system was also used to simulate the wavefront patterns of the IOLs in decentered and tilted situations. The amounts of decentration and tilt chosen for this study were based on the study of Eppig et al24 who showed that the mean values of decentration and tilt reported in the literature are 0.30±0.16 mm and 2.62±1.14°, respectively.
In regards to decentration and tilt, visual quality decreased for both low and high hyperopic groups, with the exception of the Akreos Adapt IOL for the low hyperopic group. In these cases, the lens obtained a constant visual quality over the 2° of tilt, and these results are comparable to those obtained with the lens in the centered position. For all misaligned positions studied, 0.4 mm of decentration provided the higher decrease of visual quality in all IOLs. Among the studied IOLs, the Akreos Adapt obtained better results when it was misaligned. Partially in accordance with our results, previous studies showed that IOLs with higher negative amounts of spherical aberration are the most sensitive to misalignments.23–25 However, several differences are present between these studies and our work that should be taken in consideration. Previous studies used a model eye to assess the optical quality of the IOL as a function of misalignments whereas we used an adaptive optics visual simulator to evaluate the impact of misalignment on patient visual quality. In addition, in our study, negative increments of corneal spherical aberration due to hyperopic LASIK ablations were considered. If the impact of tilt was compared in both groups, it can be observed that for high hyperopic ablations, tilted IOLs provide a worse visual quality than for low hyperopic ablations. Therefore, these results suggest that all simulated IOLs are more robust to tilt in patients with low hyperopic ablations than high hyperopic ablations.
Thus, tilt and decentration of the IOLs in addition to corneal spherical aberration play a significant role in the potential visual performance provided by the IOLs studied. Another important factor to consider in both spherical and aspheric IOLs is the spherical aberration of the IOL itself. Different values of spherical aberration of the IOLs provoke different residual spherical aberration after surgery26 and may result in different visual outcomes. For that reason, further studies should be developed with other models and designs of IOLs (with other amounts of spherical aberration), and by doing so, other combinations of higher order aberrations will be established. We would like to note that a proper calculation of the IOL power is needed to reach emmetropia postoperatively, otherwise, different visual quality outcomes could be obtained.
Despite some limitations, we can conclude that the results obtained in our study through adaptive optics simulation suggest that in patients with a negative increment of corneal spherical aberration due to hyperopic LASIK, all simulated IOLs provide good visual quality when they are centered, with better results for lower increments of negative spherical aberration. On the other hand, the decrease in visual quality that occurs when these lenses are decentered or tilted demonstrates the importance of accurate implantation, even more so when the hyperopic ablation is higher and the increments are more negative. Among the three IOLs studied, the Akreos Adapt seems to be the most robust design to misalignments in this patient population.