Purpose
To compare contrast sensitivity, visual acuity (VA), and halos in subjects bilaterally implanted with 1 of 3 FDA-approved presbyopia-correcting intraocular lenses.
Design
Prospective, randomized, partially masked, multicenter clinical trial.
Methods
Seventy-eight subjects were randomized sequentially for bilateral implantation with the Crystalens AO (Bausch & Lomb Surgical), AcrySof IQ ReSTOR +3.0 (Alcon Laboratories), or Tecnis Multifocal (Abbott Medical Optics) lenses. Subjects were evaluated through visit 4 (4 to 6 months after surgery) with the following monocular and binocular assessments: high- and low-contrast VA, contrast sensitivity without glare, halos or starbursts, defocus curves, optical scatter, retinal point spread function, and safety.
Results
The Crystalens AO and ReSTOR +3.0 demonstrated better monocular and binocular contrast sensitivity without glare at low to mid spatial frequencies compared with the Tecnis Multifocal lens. Binocular uncorrected distance VA was not significantly different between the 3 lenses. The Crystalens AO had significantly better binocular low-contrast distance-corrected VA than the ReSTOR +3.0 and better mean monocular low-contrast DCVA than the Tecnis Multifocal lens. The Crystalens AO demonstrated significantly better monocular and binocular uncorrected and distance-corrected intermediate VA than the ReSTOR +3.0 or Tecnis Multifocal lenses. The ReSTOR+3.0 lens had significantly better monocular and binocular uncorrected and distance-corrected near VA tested at 40 cm compared with the Crystalens AO and Tecnis Multifocal lens. The Crystalens AO elicited significantly less halos than the Tecnis Multifocal lens and less optical scatter than the ReSTOR +3.0 or Tecnis Multifocal lens.
Conclusions
The Crystalens AO had statistically better uncorrected intermediate VA and distance-corrected intermediate VA than the ReSTOR +3.0 or Tecnis Multifocal lenses and fewer photic phenomenon than the Tecnis Multifocal lens. Both multifocals had better distance-corrected near VA and uncorrected near VA than the Crystalens AO. These findings may guide intraocular lens selection for individual patients seeking to optimize vision at specific vergences or lighting conditions.
Included among the increasing array of intraocular lens (IOL) options are lenses designed to increase through-focus after cataract extraction. The optimal goal is to restore distance, intermediate, and near visual function. Presbyopia-correcting multifocal or accommodating IOLs provide increased spectacle independence with an extended range of functional vision compared with monofocal non–presbyopia-correcting IOLs.
Each type of presbyopia-correcting IOL relies on a different optical strategy, which is associated with varied strengths, weaknesses, and side effects. Multifocal IOLs are designed with diffractive or combined apodized diffractive-refractive mechanisms that focus light at multiple foci in attempt to provide vision at distance, intermediate, and near. Because the total light energy is split between the primary far and near foci, the image quality and sharpness are degraded to varying degrees depending on a number of factors, including pupil size. Although spectacle independence generally is high, not all patients successfully learn to attend to the in-focus image and disregard the superimposed out-of-focus image(s) produced by multifocal IOLs, and some patients are disturbed by associated photic phenomenon and reduced contrast sensitivity.
Accommodating IOLs have a single focal point so that light is not distributed among multiple foci. The exact mechanisms that improve through focus with accommodating IOLs remain a subject of debate. Potential mechanisms include forward axial movement of the lens and optic arching in response to contraction of the ciliary muscle and changes in vitreous pressure, and dynamic changes in spherical aberration and other higher-order aberrations in addition to defocus. However, the improvement of intermediate and near vision with accommodating IOLs may be limited by postoperative contraction and fibrosis of the capsular bag, variations in size and shape of the capsular bag and rhexis, and the impact of numerous medications that can impact ciliary body contraction.
We recently demonstrated the optical bench performance of the 3 lifestyle-enhancing IOLs most commonly used in the United States: the Crystalens AO accommodating IOL (Bausch & Lomb Surgical, Aliso Viejo, California, USA), the ReSTOR +3.0 apodized diffractive multifocal IOL (Alcon Laboratories, Fort Worth, Texas, USA), and the fully diffractive Tecnis Multifocal IOL (Abbott Medical Optics, Irvine, California, USA). Although such laboratory investigations provide relevant data about the inherent optical properties of these lenses, only studies of the real-life performance after intraocular implantation can provide clinical information that incorporates retinal and neural processing, such as contrast sensitivity, as well as visual acuity (VA) at different vergences. The aim of this prospective, randomized study was to compare contrast sensitivity, high- and low-contrast VA, optical scatter, depth of focus, visual quality metrics, and glare meter outcomes with bilateral implantation of these 3 different IOLs for the treatment of aphakia and presbyopia.
Methods
Study Design
This prospective, randomized, bilateral, subject- and assessor-masked clinical trial was conducted at 2 clinical practices in the United States ( ClinicalTrials.gov identifier, NCT01122576) from May 2010 through June 2012. Subjects were masked until study exit, and clinic personnel performing postoperative assessments were masked to treatment assignment until after the database was locked for the final analysis. The surgeon and immediate surgical team (J.S.P., M.A.Q., and J.S.) were unmasked to the treatment assignment. The study was conducted in accordance with the Health Insurance Portability and Accountability Act regulations and the tenets of the Declaration of Helsinki and was approved by the Southwest Independent Institutional Review Board (Fort Worth, Texas; study and data accumulation). After providing an explanation of the nature and possible consequences of the study, all subjects provided written informed consent.
The primary end point of the study was the monocular mesopic contrast sensitivity without glare at 5 spatial frequencies (1.5, 3, 6, 12, and 18 cycles/degree [cpd]) at 1 to 2 months and 4 to 6 months after surgery. Secondary end points included uncorrected distance, intermediate (at 32 inches), and near (at 16 inches) VA; high-contrast corrected and uncorrected distance, intermediate, and near VA; distance-corrected low-contrast acuity; severity of halos and starbursts; defocus curves, objective optical scatter index, and retinal point spread function metrics; and incidence of adverse events (AEs).
Eligibility Criteria
The clinical trial included 78 subjects (156 eyes) with bilateral age-related cataracts who were 40 years of age or older at the time of enrollment. Included subjects had clear intraocular media (except for cataract), a potential for corrected distance VA of 20/32 or better; had a mesopic (3 cd/m 2 ) pupil diameter of more than 5.0 mm; had an intact centered capsulorrhexis, intact posterior capsule, and no zonular rupture at the time of surgery; had 1.25 diopters (D) or less of preoperative corneal astigmatism; and required a spherical lens power of 10.00 to 33.00 D in each eye. Subjects were excluded if either eye had corneal pathologic features potentially affecting topography; an increased risk of zonular rupture because of conditions such as pseudoexfoliation syndrome; any active inflammation or edema of the cornea such as keratitis, keratoconjunctivitis, or keratouveitis; uncontrolled glaucoma; previous retinal detachment; rubella, bilateral congenital, traumatic, complicated, or polar cataract; marked microphthalmos or aniridia; previous corneal surgery; irregular corneal astigmatism; optic atrophy; iris neovascularization; chronic use of systemic steroids or immunosuppressive medications; and any ocular pathologic features or disorders causing potential acuity of 20/32 or worse (i.e., visually significant diabetic retinopathy, amblyopia, clinically significant retinal pigment epithelium, or macular changes) or potential acuity losses to 20/32 or worse (e.g., macular degeneration).
Randomization and Surgery
All subjects underwent small-incision phacoemulsification cataract surgery with a continuous and intact circular curvilinear tear capsulorrhexis. Before surgery, subjects were randomized sequentially by study personnel 1:1:1 to each of 3 study arms (n = 26 subjects/group) and bilaterally implanted with Crystalens AO (models AT-50AO or AT-52AO), AcrySof IQ ReSTOR +3.0 (model SN6AD1), or Tecnis Multifocal (model ZMA00) IOLs. Block randomization was used to balance treatment groups over time. A minimum of 2 weeks was required between the date of surgery of the first (or primary) eye and implantation of the second (or fellow) eye. The first eye to be operated on was chosen at the discretion of the surgeon. For the Crystalens AO, the dominant eye of each subject was targeted between plano and −0.25 D. If the dominant eye could not be determined, the eye targeted between plano and −0.25 D was at the surgeon’s discretion. The nondominant eye could be targeted between emmetropia and −0.5 D, at the surgeon’s discretion and after discussion with the subject after the first eye surgery. For subjects implanted with the Tecnis Multifocal lens, each eye was targeted for plano to −0.25 D. For subjects implanted with the ReSTOR +3.0 lens, emmetropia was targeted in both eyes. Personalized A-constants were permitted.
The Crystalens AO is a multipiece, hinged-plate haptic, accommodating IOL (5-mm optic, 11.5 or 12 mm overall length loop tip to tip for AT-50AO and AT-52AO, respectively). The dioptric powers range from +17.00 to +33.00 D for the AT-50AO model and +10.00 to +16.50 D for the AT-52AO model. This biconvex lens manufactured from a third-generation silicone material called Biosil (Nusil Corporation, Carpinteria, California, USA) has flexible hinged-plate haptics designed to allow movement or changes in the position and shape of the lens optic in response to accommodative effort.
The AcrySof IQ ReSTOR +3.0 lens is an apodized diffractive multifocal IOL (6.0-mm optic; 13-mm length) with dioptric powers ranging from +6.00 to +34.00 D, but for the purpose of this study, the dioptric powers used were +10.00 to +33.00 D. The IOL has a +3.00-D add. It has a central 3.6-mm apodized optic region of 9 concentric diffractive zones on the anterior surface, with a gradual reduction in diffractive step heights from center to periphery. This results in an energy continuum for light to be directed at 2 primary foci, distance and near.
The Tecnis Multifocal lens is a diffractive aspheric IOL (6-mm equiconvex square-edged optic, 13-mm length) with dioptric powers ranging from +5.00 to +34.00 D. The near power represents a +4.00-D add. For the purpose of this study, the dioptric powers used were +10.00 to +33.00 D. The diffraction pattern consists of 32 concentric rings with equal step heights. This lens is a second-generation acrylic multifocal IOL with a diffractive zone across the posterior surface of the 6.0-mm optic that provides a far and a near focal point that are 4.0 D apart.
Study Procedures
The following examinations were performed before surgery: determination of ocular dominance, keratometry, optical biometry or immersion ultrasonography, axial length, mesopic (3 cd/m 2 ) and photopic (85 cd/m 2 ) pupil diameter using the Colvard Pupillometer (Oasis, Glendora, California, USA), manifest refraction, monocular and binocular uncorrected distance VA (UDVA) and corrected distance VA (CDVA), mesopic contrast sensitivity with and without glare, slit-lamp examination, dilated fundus evaluation, retinal or potential acuity meter, and intraocular pressure by Goldmann applanation tonometry. Uncorrected and corrected distance (20 feet), intermediate (32 inches), and near (16 inches) VAs, low-contrast distance VA, and monocular and binocular corrected distance contrast sensitivities were measured using the Optec 6500/6500P Vision Tester (Stereo Optical Co., Chicago, Illinois, USA). The unit was calibrated for distance vision at 6 m, intermediate vision at 80 cm, and near vision at 40 cm.
Optical scatter via the objective scatter index (OSI) and optical quality metrics derived from the retinal point spread function (PSF) were assessed using the Optical Quality Analysis System II (OQAS; Visiometrics, Terrassa, Spain), which provides an objective assessment of visual quality achieved by subjects. The OQAS directly measures the PSF of the optical system of the eye via a double-pass technique that records images of a point source object after reflection off the retina and double-pass through the ocular media. The asymmetric configuration of the double-pass technique allows the capture of possible asymmetries in the retinal images, which can be derived from the composite of diffraction, aberrations, and scatter.
The subjects’ optical quality and ocular scattering were measured using the OQAS at the best focus position to achieve the best possible optical quality. The subjects’ refractive errors were corrected during these measurements: the spherical defocus error was corrected automatically by the double-pass system and astigmatism of more than 0.50 D was corrected with an external cylindrical lens.
The double-pass system provided the retinal image for each analyzed eye, from which the monochromatic modulation transfer function (MTF) was computed. This function represents the loss of contrast produced by the optics of the eye as a function of spatial frequency. The system also provides several simpler parameters that are related to the MTF profile: the MTF cutoff, the 2-dimensional Strehl ratio (i.e., the ratio between the areas under the MTF curve of the measured eye and that of the aberration-free eye), and the OQAS values at 100%, 20%, and 9% contrast. The MTF cutoff is calculated as that corresponding to a 0.01 MTF value, because there is background noise in the MTF profile that is computed from the real recorded double-pass image. This value is the cutoff point of the MTF on the x-axis in cycles per degree, representing the point at which the spatial frequency is maximal. The PSF width at 50% and 10% of PSF height also was determined.
The OSI is the ratio between the integrated light in the periphery and the central peak of the double-pass image. The higher the OSI value, the higher the level of intraocular scattering. The OSI values range from 0 (no scatter) to a maximum of 25 (with <1 associated with no cataract and 2 associated with Lens Opacities Classification System grade II).
Follow-up examinations were conducted at 1 day (visit 1), 7 to 14 days (visit 2), 1 to 2 months (visit 3), and 4 to 6 months (visit 4) after IOL implantation. Postoperative assessments included evaluation of keratometry; mesopic and photopic pupil diameter; monocular and binocular mesopic contrast sensitivity with and without glare; monocular and binocular UDVA, uncorrected intermediate VA (UIVA), and uncorrected near VA (UNVA); monocular and binocular CDVA, distance corrected intermediate VA (DCIVA), and distance corrected near VA (DCNVA); manifest refraction; slit-lamp examination; undilated monocular and binocular determinations of halo and starburst using a 5-point integer scale on the Glarometer (Gulden Ophthalmics, Elkins Park, Pennsylvania, USA); defocus curves; intraocular pressure; and AEs.
Defocus testing was performed in mesopic conditions (approximately 60 to 80 cd/m 2 ) using a phoropter or trial frame with a contrast Early Treatment Diabetic Retinopathy Study chart at 4.0 m. The distance-corrected manifest refraction was used to designate the 0 baseline. A defocus of +4.00 D spherical correction from the manifest refraction was set and the Snellen acuity was recorded. Positive spherical power was decreased in 0.50-D increments with Snellen VA recorded for each change in correction until only the manifest refraction remained. A defocus of −4.00 D spherical correction then was set from the manifest refraction, and the Snellen VA at this refraction was recorded. If the subject could still see target at −4.00 D sphere, negative sphere was added until the target blurred, and the Snellen VA at this refraction was recorded. Negative spherical power was decreased in 0.50-D increments with Snellen VA recorded at each change in correction until only the manifest refraction remained.
Statistical Methods and Analyses
A sample size of 32 in each group would have 90% power to detect a difference in means with a 0.0025 2-sided α risk level. Because this was a bilateral study, the number of subjects required was 16 (one half of the number of eyes). However, contrast sensitivity was expected to be correlated between eyes within subjects, so the analyzable sample size had to be increased to 23 subjects. Allowing for up to 10% dropouts, the number of subjects to be implanted was 26 per treatment group.
Continuous monocular primary or secondary end points were contrasted with pairwise comparisons by visit using mixed models including the fixed effect of lens type and the random effect of subject nested within treatment group. Continuous or ordinal binocular (or subject-level) end points were compared pairwise by visit using Wilcoxon-Mann-Whitney rank-sum tests and corresponding Hodges-Lehman confidence intervals. Ordinal monocular end points were compared by visit using generalized estimating equations assuming multinomial distributions and cumulative logit link functions. Exact Clopper-Pearson 95% confidence intervals for the number of eyes with at least 1 ocular AE are presented. Means and standard deviations (SD) are presented, and statistical significance was P < .05.
Results
Subject disposition is reported in Figure 1 . Demographic data are summarized in Table 1 . There were no significant differences in the demographic features among all 3 cohorts. The mean ± SD photopic and mesopic pupil diameter for the Crystalens AO cohort was 3.21 ± 0.56 mm and 6.09 ± 0.70 mm, respectively. For the ReSTOR +3.0 cohort, the mean ± SD pupil diameter was 3.24 ± 0.71 mm photopic and 6.17 ± 1.27 mm mesopic, and for the Tecnis Multifocal cohort, mean ± SD pupil diameter was 3.41 ± 0.61 mm photopic and 6.43 ± 0.84 mm mesopic. The mean photopic and mesopic pupil sizes between the 3 groups were not statistically different.
| Crystalens AO (n = 26) | ReSTOR +3.0 (n = 26) | Tecnis Multifocal (n = 26) | P Value | |
|---|---|---|---|---|
| Age (y) | ||||
| Mean (SD) | 62.8 (5.9) | 64.2 (7.0) | 63.0 (8.8) | .776 |
| Minimum, maximum | 52, 73 | 48, 76 | 43, 81 | |
| 95% CI | 60.5 to 65.2 | 61.3 to 67.0 | 59.5 to 66.6 | |
| Group, no. (%) | ||||
| 40 to 49 | 0 | 1 (3.8) | 1 (3.8) | |
| 50 to 59 | 10 (38.5) | 5 (19.2) | 8 (30.8) | |
| 60 to 69 | 13 (50.0) | 12 (46.2) | 12 (46.2) | |
| 70 to 79 | 3 (11.5) | 8 (30.8) | 4 (15.4) | |
| ≥80 | 0 | 0 | 1 (3.8) | |
| Gender | ||||
| Male, no. (%) | 9 (34.6) | 11 (42.3) | 7 (26.9) | .507 |
| 95% CI | 16.3 to 52.9 | 23.3 to 61.3 | 9.9 to 44.0 | |
| Female, no (%) | 17 (65.4) | 15 (57.7) | 19 (73.1) | |
| 95% CI | 47.1 to 83.7 | 38.7 to 76.7 | 56.0 to 90.1 | |
| Race, no. (%) | ||||
| White | 26 (100) | 26 (100) | 26 (100) | |
| Ethnicity, no. (%) | ||||
| Not Hispanic and not Latino | 26 (100) | 26 (100) | 26 (100) |
Quantity of Vision Metrics
High-contrast uncorrected distance, intermediate, and near visual acuity
Figure 2 shows the comparative cumulative monocular and binocular high-contrast UDVA at visit 4 (4 to 6 months after surgery) and the mean UDVA. The ReSTOR +3.0 cohort had significantly better monocular mean UDVA compared with the Crystalens AO cohort at visit 3 (1 to 2 months after surgery; P < .001) and visit 4 ( P = .008) and compared with the Tecnis Multifocal cohort at visit 3 ( P = .035). However, data relating to the dominant eye at visit 4 showed that the mean monocular high-contrast UDVA was not significantly different between the Crystalens AO cohort and either the ReSTOR +3.0 cohort ( P = .053) or the Tecnis Multifocal cohort ( P = .430). These results are consistent with more myopic targeting of the nondominant eye within the Crystalens AO cohort. The ReSTOR +3.0 and Tecnis Multifocal cohorts had significantly better binocular UDVA ( P = .009 and P = .039, respectively) compared with the Crystalens AO cohort at visit 3, but there were no statistically significant differences at visit 4 (Crystalens vs ReSTOR +3.0, P = .569; Crystalens vs Tecnis Multifocal, P = .245; and ReSTOR +3.0 vs Tecnis Multifocal, P = .755).
Monocular and binocular mean UIVA results, with testing at 80 cm, show that the Crystalens AO was significantly better at visits 3 and 4 ( Figure 3 ) compared with the ReSTOR +3.0 IOL ( P < .001 for both visits) and Tecnis Multifocal IOL ( P = .003 binocularly at visit 3; P < .001 monocularly visits 3 and 4 and binocularly visit 4). No statistically significant differences were found for both monocular and binocular UIVA results between the ReSTOR +3.0 and Tecnis Multifocal groups ( Figure 3 ).
The ReSTOR +3.0 cohort showed significantly better monocular and binocular UNVA, with testing at 40 cm, at visits 3 and 4 ( Figure 4 ) compared with the Crystalens AO ( P < .001 at both visits) and Tecnis Multifocal ( P < .001 at both visits) cohorts. There were no significant differences in UNVA between the Crystalens AO and Tecnis Multifocal cohorts, with the exception of the mean binocular UNVA at visit 4 in favor of the Tecnis Multifocal lens ( P = .022).
High-contrast distance-corrected visual acuity
Intermediate and near visual acuities were tested using the distance correction to eliminate the potential pseudoaccommodative effects of residual myopia and cylinder. At visit 4, monocular mean high-contrast CDVA was not significantly different between groups: mean logarithm of the minimal angle of resolution VA ± SD for the Crystalens AO cohort, −0.100 ± 0.066 (Snellen equivalent, 20/16); ReSTOR +3.0, −0.096 ± 0.106 (Snellen equivalent, 20/16); and Tecnis Multifocal, −0.075 ± 0.088 (Snellen equivalent, 20/20). Likewise, at visit 4, binocular mean high-contrast CDVA was similar between groups: Crystalens AO, −0.136 ± 0.049 (Snellen equivalent, 20/15); ReSTOR +3.0, −0.137 ± 0.058 (Snellen equivalent, 20/15); and Tecnis Multifocal, −0.117 ± 0.098 (Snellen equivalent, 20/15). For DCIVA, the Crystalens AO was significantly superior to the ReSTOR +3.0 and Tecnis Multifocal lenses monocularly and binocularly ( P < .001 for both; Figure 5 ). For monocular and binocular DCNVA, the ReSTOR +3.0 and Tecnis Multifocal lenses performed significantly better than the Crystalens AO, and the ReSTOR +3.0 performed significantly better than the Tecnis Multifocal lens ( P < .001 for all; Figure 6 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
