Gian Marco Tosi MD Dipartimento di Scienze Oftalmologiche e Neurochirurgiche Università di Siena Viale Bracci 1 53100 Siena Italy Tel: + 39 0577 585660 Fax: + 39 0577 586162 Email: firstname.lastname@example.org
Purpose: To compare differences in subjective glare and spherical aberration between five foldable intraocular lenses (IOLs) made of different materials and to different designs.
Methods: This prospective study comprised 175 cataract patients who underwent phacoemulsification and were randomized to receive one of five types of foldable IOL (AcrySof® MA30BA, Alcon; Sensar® AR40, AMO; AcrySof® SA30AL, Alcon; Sensar® AR40e, AMO, and Tecnis® Z9000, Pharmacia & Upjohn). All patients received a questionnaire investigating the incidence of subjective photic phenomena. Two months postoperatively, we collected data regarding subjective glare and evaluated pupil size, visual acuity and wavefront aberration of the cornea and eye.
Results: With regard to difficulty in performing ordinary activities under different light conditions and light and dark adaptation, the difference between the groups was not significant (p > 0.05, chi-squared test). With respect to difficulty in driving at night, the MA30BA group had a significant higher incidence of photic phenomena than the SA30AL, AR40e and Z9000 groups (p < 0.05, chi-squared test). Wavefront measurements revealed a significant difference between the Z9000, AR40e and SA30AL groups, which showed the lowest values, and the MA30BA group, which showed the highest value (p < 0.05, anova with Tamhane posthoc test).
Conclusions: New generation IOLs such as the Pharmacia Z9000, AMO AR40e and AcrySof® SA30AL have a lower incidence of glare and spherical aberrations; however, their impact on future IOL design should be conditioned by further data, especially regarding posterior capsule opacification.
Cataract removal with intraocular lens (IOL) implantation is nowadays considered the most successful procedure in modern medicine worldwide; increasingly, this procedure is being used for refractive purposes (Kohnen 2003). Although the achievement of uncorrected 20/20 visual acuity (VA) is satisfactory, much effort has been made to further improve the optical quality provided by IOLs (Uchio et al. 1995a; Holladay et al. 1999, 2002; Mester et al. 2003). Patients who receive IOLs after cataract surgery occasionally experience some degree of glare, arcs or halos. Such dysphotopsias, which remain among the most common sources of dissatisfaction in an otherwise satisfied pseudophakic population, can result from posterior capsule opacification (PCO), postoperative mydriasis or IOL-related factors (Koch et al. 1986; Uchio et al. 1995a; Tester et al. 2000; Erie et al. 2001; Erie & Bandhauer 2003).
Moreover, the recent introduction of wavefront sensing in excimer corneal refractive surgery has increased interest in the study and correction of spherical aberration introduced by IOLs. Spherical aberration is a property of spherical lenses, which bend peripheral rays more or less strongly so that these rays cross the optical axis in front of (positive spherical aberration) or behind (negative spherical aberration) the paraxial rays (Packer et al. 2002; Kohnen 2003). Spherical IOLs have been shown to induce positive spherical aberrations, thus contributing to a decline in image quality after cataract surgery (Uchio et al. 1995a; Holladay et al. 2002; Mester et al. 2003).
The possible common pathogenesis, together with the increasing number of new generation foldable IOLs, induced us to verify the occurrence of postoperative subjective glare and spherical aberration and their correlation with five different IOL types.
Material and Methods
This prospective randomized study comprised 175 consecutive cataract patients (175 eyes) referred to the Department of Ophthalmology and Neurosurgery at the University of Siena from September to December 2002. The study design was approved by the Ethical Committee of the University of Siena and the Siena University Medical Hospital and was conducted in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki, as renewed in 2000. All patients gave written informed consent prior to their inclusion in the study.
Inclusion criteria were age between 50 and 80 years, preoperative astigmatism up to 1 dioptre (D) and IOL power between + 18 D and + 24 D. Exclusion criteria were other ocular pathologies, previous anterior and posterior segment surgery and intraoperative and postoperative complications.
All patients underwent a standardized small incision phacoemulsification with foldable IOL implantation in the capsular bag. They were randomized into groups of 35 (35 eyes) to receive one of the following foldable IOLs: AcrySof® MA30BA (Alcon Laboratories, Inc., Fort Worth, TX, USA); Sensar® AR40 (Advanced Medical Optics, Inc. (Amo), Santa Ana, CA, USA); AcrySof® SA30AL (Alcon); Sensar® AR40e (AMO), or Tecnis® Z9000 (Pharmacia & Upjohn Co. Kalamazoo, MI, USA).
All operations were performed by the same surgeon (AC). In all cases, the Alcon Legacy phacoemulsification machine was used. All patients received peribulbar injection or topical anaesthesia before surgery. A 2.75-mm wide, self-sealing superior sclerocorneal tunnel was created. A 5–5.25-mm continuous curvilinear capsulorhexis was made with Caporossi's forceps (Janach) and the nucleus removed by the divide-and-conquer method. For insertion, all IOLs were implanted with the Universal Caporossi folder and holder (Janach), except for AMO AR40 and AR40e IOLs, which were implanted with the AMO Unfolder Sapphire. The scleral wound was not sutured.
The following investigations were performed 2 months postoperatively: assessment of the anterior and posterior segments including IOL decentration and tilt, pupil size, VA, spherical aberration of the cornea and eye, and subjective glare.
Visual acuity was measured in all patients under the same ambient conditions; pupil size was measured with a corneal topography unit (Eye-Top; Costruttori Strumenti Oftalmici [CSO], Florence, Italy).
Wavefront aberration of the cornea was evaluated preoperatively and 2 months postoperatively using a corneal topography unit (Eye-Top; CSO). The corneal wavefront spherical aberration (Z[4.0]) was calculated by analysing the corneal elevation height data of the Zernike coefficients and a comparison between preoperative and postoperative data was performed. Wavefront spherical aberration (Z[4.0]) of the whole eye was evaluated after pharmacological dilatation of the pupil and calculated for a 5.0-mm pupil using a Hartman−Shack wavefront sensor (Zywave; Bausch & Lomb, Rochester, NY, USA). The principles associated with this technique have been explained previously (Liang et al. 1994). A patient questionnaire asking for information on the incidence of subjective photic phenomena was given to all patients before surgery. This has been described earlier (Casprini et al. 2002) (Fig. 1).
Differences between groups were compared using one-way analysis of variance (anova) for age, pupil size and aberration data. Visual acuity was compared using the Kruskall−Wallis non-parametric test. For statistically significant outcomes, two-by-two group differences were evaluated using a suitable posthoc test; Bonferroni and Tamhane tests were chosen for equal or unequal variance within the groups, respectively. Homogeneity of variance within groups was preliminarily tested by Levene statistics. Finally, the differences between the numbers of positive answers to the questionnaire were compared using the chi-squared test. P-values < 0.05, corresponding to a significance level > 95%, were considered significant. Non-significant p-values between 0.05 and 0.1 were also reported.
All statistical analyses were performed using spss.
Each group of 35 patients randomly received one of the following IOLs: AcrySof® MA30BA (Alcon); Sensar® AR40 (AMO); AcrySof® SA30AL (Alcon); Sensar® AR40e (AMO), or Tecnis® Z9000 (Pharmacia & Upjohn Co.). None of the 175 recruited patients dropped out of the study. At the 2-month follow-up visit, none of the patients showed any anterior or posterior capsule opacification. Neither IOL decentration nor IOL tilting was observed. No statistically significant differences were found between subgroups in the evaluation of pupil sizes with the corneal topography unit at the 2-month follow-up visit (p > 0.1, anova test) (Table 1).
Table 1. Mean pupil diameter and mean BCVA at the 2-month follow-up visit.
BCVA = best corrected visual acuity.
Pupil size (mm ± SD)
2.41 ± 0.41
2.54 ± 0.47
2.38 ± 0.42
2.53 ± 0.39
2.45 ± 0.49
BCVA ± SD
0.93 ± 0.074
0.92 ± 0.08
0.91 ± 0.08
0.91 ± 0.076
0.96 ± 0.061
The mean age of the patients was 72.25 ± 9.2 years (SD) in the AcrySof MA30BA group, 74.6 ± 9.4 years (SD) in the Sensar AR40 group, 70.61 ± 4.3 years (SD) in the AcrySof SA30AL group, 73.8 ± 7.1 years (SD) in the Sensar AR40e group, and 72.9 ± 8.8 years (SD) in the Pharmacia Z9000 Technis group. There were no statistically significant differences bet-ween groups in terms of age (p > 0.1, anova test).
Two months after surgery, no statistically significant difference between IOL groups emerged in terms of best corrected VA (p > 0.1, Kruskall−Wallis test) (Table 1).
Likewise, neither preoperative nor postoperative corneal aberration values nor their comparison showed any statistically significant difference between IOL groups (p > 0.1, anova test) (Table 2; Fig. 2).
Table 2. Pre- and postoperative corneal aberration data.
Z2 µm ± SD
Z3 µm ± SD
Z4 µm ± SD
Z5 µm ± SD
0.27 ± 0.09
0.5 ± 0.22
0.31 ± 0.15
0.22 ± 0.1
0.19 ± 0.03
0.47 ± 0.2
0.38 ± 0.07
0.26 ± 0.12
0.41 ± 0.21
0.62 ± 0.19
0.32 ± 0.09
0.35 ± 0.25
0.28 ± 0.10
0.64 ± 0.12
0.29 ± 0.06
0.32 ± 0.07
0.22 ± 0.09
0.46 ± 0.13
0.31 ± 0.11
0.18 ± 0.06
0.43 ± 0.31
0.54 ± 0.21
0.45 ± 0.19
0.48 ± 0.25
0.3 ± 0.23
0.65 ± 0.28
0.47 ± 0.16
0.27 ± 0.04
0.55 ± 0.27
0.66 ± 0.20
0.4 ± 0.15
0.33 ± 0.08
0.31 ± 0.02
0.59 ± 0.19
0.36 ± 0.05
0.36 ± 0.12
0.3 ± 0.09
0.54 ± 0.24
0.43 ± 0.13
0.23 ± 0.1
At 2 months postoperatively, mean spherical aberration coefficient (Z[4.0]) of the whole eye for a 5.0-mm pupil was 1.01 µm ± 0.85 (SD) in the MA30BA group, 0.75 µm ± 0.71 in the AR40 group, 0.51 µm ± 0.48 in the SA30AL group, 0.45 µm ± 0.39 in the AR40e group, and 0.4 µm ± 0.35 in the Z9000 group (Table 3; Fig. 3). A statistically significant difference between the SA30AL, AR40e and Z9000 groups, which showed the lowest values, and the MA30BA group, which showed the highest value, was found (p < 0.05, anova with Tamhane posthoc test; unequal variance verified by Levene statistics). No statistically significant differences were found between the other examined subgroups (p > 0.05, anova with Tamhane posthoc test). A p-value equal to 0.072 was found when comparing the AR40 and AR40e groups and a p-value equal to 0.055 resulted from the comparison between the AR40 and Z9000 groups.
Table 3. Mean aberration data of the whole eye for a 5.0-mm pupil.
Z2 µm ± SD
Z3 µm ± SD
Z4 µm ± SD
Z5 µm ± SD
0.8 ± 0.40
0.78 ± 0.65
1.01 ± 0.85
0.75 ± 1.02
0.44 ± 0.56
0.81 ± 0.72
0.75 ± 0.71
0.39 ± 0.50
0.3 ± 0.24
0.45 ± 0.30
0.51 ± 0.48
0.35 ± 0.47
0.33 ± 0.45
0.77 ± 0.95
0.45 ± 0.39
0.23 ± 0.33
0.28 ± 0.82
0.4 ± 0.34
0.4 ± 0.35
0.12 ± 0.21
All questionnaires were filled out and returned 2 months postoperatively (Fig. 4).
Regarding questions 1, 2, 3 and 4 about trouble performing ordinary activities under different light conditions, the five IOL groups showed similar percentages of positive answers: 20.5% in the MA30BA group, 19% in the AR40 group, 13.7% in the SA30AL group, 11.5% in the AR40e group and 13.7% in the Z9000 group. The difference between the groups was not significant (p > 0.1, chi-squared test).
Regarding question 5 about light adaptation, the MA30BA group showed 25% positive answers, the AR40 group 19%, the SA30AL group 13%, the AR40e group 13% and the Z9000 group 19%. The difference between the groups was not significant (p > 0.1, chi-squared test).
Regarding question 6 about dark adaptation, the MA30BA group showed 39% positive answers, the AR40 group 25%, the SA30AL group 16%, the AR40e group 16% and the Z9000 group 16%. The difference between the groups was not significant (p > 0.1, chi-squared test).
Answers to questions 7, 8 and 9 about difficulties in driving at night showed a percentage of positive answers of 47% in the MA30BA group, 33% in the AR40 group, 19% in the SA30AL group, 15% in the AR40e group and 13% in the Z9000 group. The MA30BA group had a significantly higher incidence of photic phenomena while driving at night than the SA30AL, AR40e and Z9000 groups (p < 0.05, chi-squared test). No statistically significant differences were found between the other examined subgroups (p > 0.1, chi-squared test).
Although phacoemulsification with foldable IOL implantation yields outstanding results, some patients complain of pseudophakic dysphotopsia. Sometimes these symptoms are so bothersome as to require IOL exchange (Forbowitz et al. 2000). Moreover, in a recent survey, Mamalis (2002) showed that glare and optical aberrations are among the most common reasons for removing a foldable IOL.
Comparing IOLs with different anterior and posterior optic curvature, Uchio et al. (1995a) found that glare and spherical aberration seem to have a similar tendency. These data have been indirectly confirmed by studies conducted on glare and spherical aberration separately. In fact, more pronounced anterior optic curvature or an equi-biconvex design has been shown to reduce pseudophakic dysphotopsia (Erie et al. 2001) and to reduce spherical aberrations (Atchison et al. 1990, 1991; Uchio et al. 1995a). Moreover, the major changes that occur in the lens with age can result in a loss of negative compensation of the positive spherical aberrations introduced by the cornea, thus increasing not only ocular aberrations but also optical side-effects such as glare and halos (Holladay et al. 2002).
In the present study, we aimed to verify the occurrence of postoperative subjective glare and spherical aberration in pseudophakic patients implanted with five different IOL types. Reducing the spherical aberration of the whole eye has previously been shown to improve optical quality for pseudophakic patients (Holladay et al. 2002; Mester et al. 2003). Consequently, in our analysis of the aberrations of the whole eye, IOL groups were compared using only the mean spherical aberration coefficient (Z[4.0]).
Our results point in the same direction as those mentioned above. In fact, the SA30AL, AR40e and Z9000 groups, compared to the MA30BA group, showed not only significantly reduced subjective glare but also reduced spherical aberrations. Our data seem to confirm the importance of anterior surface optic shape and curvature in glare and spherical aberration formation. In fact, the modified prolate (flatter curvature in the periphery) anterior optic shape of the Z9000 IOL has been previously shown to reduce spherical aberrations of the eye after cataract surgery by compensating for the positive spherical aberration (Holladay et al. 2002; Mester et al. 2003), and to improve contrast sensitivity under both mesopic and photopic conditions (Packer et al. 2002). Moreover, Erie et al. (2001) reported that IOLs with a reduced anterior radius of curvature can minimize surface reflections: the SA30AL and the AR40e IOLs have anterior radii of curvature of 21 mm and 15 mm, respectively, while the MA30BA has an anterior radius of curvature of 32 mm. In addition, the IOL refraction index appears to be important. Low RI has been shown to reduce glare (Erie et al. 2001, 2003); both the Z9000 and AR40e have a low RI (1.46 and 1.47, respectively). However, surprisingly, notwithstanding the same high RI (1.55), the SA30AL IOL gave a significantly better performance than the MA30BA IOL.
Although not statistically significant, the AR40e IOL has shown slightly better optical performances than the AR40 IOL. These two IOLs differ only in optic edge profile: the AR40e IOL has round anterior and square posterior edges while the AR40 IOL has round anterior and posterior edges. Holladay et al. (1999) showed that a reduction in glare phenomena can be achieved by changing the optic edge from sharp to round. On the other hand, Nishi & Nishi (1999) reported that a sharp-edged optic design causes less PCO than round-edged optic. Although the AR40e IOL was designed specifically in order to reduce both PCO and glare, the difference in terms of postoperative glare between this IOL group and the AR40 group was unexpected. From our study, the pathogenesis of edge glare is even more difficult to assess if we analyse the edge profiles of the SA30AL, Z9000, AR40e and MA30BA. The SA30AL has a frosted edge, the AR40e has round anterior and square posterior edges, while both the MA30BA and Z9000 IOLs have square anterior and posterior edges.
Another IOL characteristic that seems worthy of consideration is the optic diameter. The role of optic diameter in the genesis of postoperative glare is disputed. Some authors have shown (Uchio et al. 1995a) or postulated (Ellis 2001) better results with greater optic diameter, while others (Davison 2000; Forbowitz et al. 2000) have not. In our study, the best results were obtained with the AR40e and Z9000 IOLs (optic diameter of 6 mm). However, no significant differences were found between the AR40e, the Z9000 and the SA30AL, notwithstanding the fact that the latter IOL has an optic diameter of 5.5 mm. Consequently, according to the analysis of our data, optic diameter, edge profile and RI seem to act not as independent factors in glare formation, but in combination with other IOL characteristics.
The remaining IOL evaluated in this study (the AR40) showed intermediate results, and thus did not significantly differ from the other IOLs. To our knowledge, no comparative evaluation in terms of spherical aberration and glare between the AR40, on the one hand, and the AR40e and Z9000 IOLs, on the other, has been made up to now. However, recent studies have shown the AR40 IOL to have lower postoperative glare than the MA30BA IOL (Casprini et al. 2002) and lower external and internal reflections than the SA30AL IOL (Erie et al. 2003).
Although the relationships among IOL-related factors in the induction of glare and spherical aberrations are complex and a more numerous case series is certainly needed, our results seem to favour the new generation IOLs. However, there are some issues to consider before drawing definitive conclusions.
The questionnaire could have increased the subjects' perception of photic phenomena and we cannot exclude the notion that the discomfort they experienced was only transient. In fact, Arnold (1994) reported that dysphotopsia is resolved quickly and spontaneously in two-thirds of subjects, as was also shown by our previous findings.
The spherical aberration related to the IOL was measured indirectly. In fact, the results obtained using the CSO Eye-Top corneal topography unit are not comparable to those obtained with the Hartman−Shack wavefront sensor. Preoperative and postoperative corneal aberration measurements were only used to demonstrate that the subgroups analysed were highly homogeneous, in order to ensure that any differences in results could be attributed to IOL type. Moreover, aberrations induced by the posterior corneal surface were not measured. We hope that, in the near future, the degree to which a particular IOL is responsible for generating spherical aberrations can be measured directly.
As our observation was limited to a 2-month follow-up period, we cannot assess whether posterior capsule opacification would have occurred and whether it would have influenced our results. In fact, the MA30BA IOL, which showed the highest incidence of glare and spherical aberrations, has been demonstrated to cause a very low rate of PCO due to its material and design (Ursell et al. 1998; Nishi & Nishi 1999). To our knowledge, the Z9000 (silicone IOL with a sharp optic edge) and AR40e (acrylic IOL with round anterior and sharp posterior optic edges) have never been tested for incidence of PCO.
An effort to verify the incidence of glare more precisely should be made. Davison (2000) showed it to be 0.2% in a large series, but it is not specified in the majority of studies. If we compare this estimate to the reported incidence of PCO (from 0.9% to 17.1% with foldable IOLs) (Apple et al. 2001), formation of which may lead to a reduction in visual acuity and the treatment of which may lead to sight-threatening complications, it is clear that further studies on new generation IOLs, especially in terms of PCO, are needed.
In recent years the study of the optical quality provided by IOLs has received ever-increasing attention. This prospective randomized study aimed to show the differences in subjective glare and spherical aberrations between five foldable IOLs. New generation IOLs such as the Pharmacia Z9000, AMO AR40e and AcrySof SA30AL® showed a reduced incidence of subjective glare and spherical aberrations. Further studies including more numerous case series, longer follow-up and, possibly, the use of instruments directly measuring spherical aberrations induced by IOLs are needed to confirm these preliminary results and to understand the exact roles of the various IOL-related factors that lead to glare and spherical aberrations.
The study was supported by the University of Siena and by the Monte dei Paschi Bank Foundation, Siena, Italy.