Diffractive bifocal IOLs divide light into two foci; in Tecnis multifocal IOLs, 41% of incoming light is used for distance vision and 41% for near vision, independent of the pupil diameter, whereas the remaining 18% is lost to higher-order scattering18,19. Diffractive trifocal IOLs divide light into three foci; moreover, PanOptix lenses are relatively unaffected regardless of pupil diameter, with 88% of the incoming light reaching the retina. 50% of this is used for distance vision, 25% for intermediate vision and 25% for near vision7. Cochener20 evaluated the postoperative visual function of bifocal IOL (ZMB00) and trifocal IOL (FineVision) and reported a better visual acuity with the trifocal IOL for intermediate vision and no difference for near vision. In our study, intermediate visual acuity was better with trifocal IOL than with bifocal IOL, but near visual acuity was better with bifocal IOL than with trifocal IOL. Tecnis multifocal IOLs have a prescription power of + 4 D and an ideal near focal length of 33 cm, and PanOptix lenses have a prescription power of + 2.17 D and + 3.25 D, with ideal focal distances of 40 cm in the near range and 60 cm in the intermediate range. The far vision was tested at 5 m, the intermediate vision was tested at 50 cm, and the near vision was tested at 30 cm for the examination of visual acuity in our study, and Tecnis multifocal IOLs had significantly better near visual acuity than PanOptix, which was partially due to the fact that their ideal near focal length is closer to the near acuity test distance. A greater percentage of spectral proportions in the near region of Tecnis multifocal IOLs may have also contributed to the better near visual acuity.
Approximately 82.2% of patients in the bifocal group and 83.7% of patients in the trifocal group were completely spectacle independent. This is consistent with the findings of other study reports on Tecnis multifocal IOLs and PanOptix IOLs, in which the percentage ranges from 82.6–92.8%21–25 and from 85.0–96.3%26,27, respectively. In this study, although the results showed that the TFNT00 was significantly superior to the ZMB00 in terms of intermediate visual acuity and that ZMB00 was significantly superior to the TFNT00 in terms of near visual acuity, there was no significant difference in spectacle independence. Good intermediate vision is necessary when using digital devices, such as computers, smartphones, and tablets, and when cooking and playing sports. Unlike the TFNT00, which is a diffractive trifocal IOL, the ZMB00, a diffractive bifocal intraocular lens, does not have spectral distribution at intermediate distances; nevertheless, there are reports that the ZMB00 provides functional intermediate vision6. Moreover, Japanese individuals, who have high percentages of myopia and presbyopia and are likely to need eyeglasses or contact lenses, can expect an improved quality of vision with the spectacle independence that they gain after TFNT00/ZMB00 insertion.
We used the CGT-1000 instrument to measure contrast sensitivity. The CGT-1000 is a dome-shaped device that measures contrast sensitivity by converting the size and luminance of the optotype, which is indicated by a double ring displayed on the monitor at a constant background luminance of 10 cd/m2 and an examination distance of 35 cm. Patients undergoing this examination are asked if they can detect any changes in the brightness contrast of a circular optotype of variable size consisting of three colored concentric circles. This device can evaluate contrast sensitivity at six sizes and 13 contrast levels, with/without glare28. Jonker et.al.29 reported that the mean mesopic contrast sensitivity was higher in the bifocal group (Acrysof IQ Restor + 3.0 bifocal IOLs) than in the trifocal group (Finevision Micro F trifocal IOLs). In contrast, Plaza-Puche et.al.30 reported that there was no significant difference in the low mesopic contrast sensitivity function among various groups, including the bifocal groups (Lentis Mplus-LS313, Acri.Lisa 366D, Acrysof ReSTOR SN6AD1) and trifocal groups (AT LISA tri 839MP, FineVision). Mojzis et.al.31 also reported that there was no significant difference in contrast sensitivity between the bifocal group (AT LISA 801) and the trifocal group (AT LISA tri 839 MP) for most of the analyzed frequencies, and Bilbao-Calabuig et.al.32 reported that there was no significant difference in contrast sensitivity between the bifocal group (ReSTOR + 2.50 and + 3.00 D) and trifocal group (FineVision). However, García-Pérez et al.4 reported that TFNT00 had slightly better contrast sensitivity compared to other trifocal IOLs, FineVision and ATLISA, and Cochener20 reported that there was no significant difference between the ZMB00 and FineVision. In our study, the result that the TFNT00 had better contrast sensitivity than the ZMB00 is consistent with the findings of these reports. In this study, contrast sensitivity was measured under corrected near vision. It can be said that the test was performed under distance optics, which produces higher visual acuity when near vision is corrected. Although TFNT00 is a quadrifocal IOL in lens design, it functions as a trifocal IOL. There is a focal length of 120 cm in addition to 40 cm and 60 cm, and this focus (120 cm) is actually redistributed to the distant focal point to enhance visual function at a distance (Enlighten optical technology)2,7. This technology allows the TFNT00 to achieve the aforementioned spectral distribution of incident light to the far end of the spectrum of 44%, which is higher than the 41% of the ZMB00. This might be one reason that the contrast sensitivity (with/without glare) of the TFNT00 was better than that of the ZMB00 in this study.
The correlation coefficients between all possible combinations of variables described in the heatmaps demonstrated that most contrast sensitivity (with/without glare) of most frequencies and UDVA/CDVA were strongly correlated in the diffractive bifocal group and that contrast sensitivity with glare (4.0/1.0 degrees) and CDVA were strongly correlated in the diffractive trifocal group (Figs. 4 and 5). In other words, CDVA is related to contrast sensitivity with glare in both the TFNT00 and ZMB00 lenses.
The NEI VFQ-25, which evaluates 25 items across 12 subscales, is one of the representative multidimensional questionnaires that have been translated into several languages and assesses the general quality of life relating to eye conditions/visual problems33. Suzukamo et al.34 validated the Japanese version of the NEI VFQ-25. The heatmap of the correlation coefficients in the diffractive bifocal IOL group demonstrated that the VFQ-25 score for Peripheral Vision and Color Vision correlated with respectively visual acuity (corrected/uncorrected) and (Fig. 4). A heatmap of the correlation coefficients in the diffractive trifocal IOL group demonstrated that the VFQ-25 score for Daytime Driving and Spectacle Dependence correlated with most ocular/internal higher-order aberrations scaled to a pupil size of 6 mm and for Color Vision correlated with CDVA/UDVA (Fig. 5). Therefore, there may be a relationship between color vision and CDVA/UDVA in TFNT00/ZMB00. One of the problems experienced by patients with multifocal IOLs is night driving, and even with the TFNT00 and ZMB00, night driving tends to be less favorable than other parameters on the Visual Function Index questionnaire (VF-14) and the NEI VFQ-25.2,6,10. Kohnen et al.2 reported that 93% of patients perceived optical phenomena such as halos (89%), glare (11%), double vision (7%), ghosting (4%), and distorted vision (4%), while 7% of patients did not report any optical phenomena with TFNT00. It is known that the neural adaptation process of vision is necessary for the brain to adapt to the different images provided by multifocal optics. Failure of this neural adaptation can result in the perception of glare, confusion, distortion, and a sense of decreased visual acuity. The typical process of neural adaptation after multifocal IOL implantation takes a minimum of three months, with maximum improvement reached one year after surgery3. In this study, the NEI VFQ-25 was performed at 10 weeks, at which point the process of neural adaptation was probably incomplete. This could be one of the reasons for the relatively low mean night driving scores in our study population.
There are some potential limitations of this study. One of them is that intermediate visual acuity was measured only at 50 cm and near visual acuity at 30 cm in the same way described in our previous study11,12. Although there has been a long-standing controversy about the examination distance35, visual acuity should ideally be measured at various distances to gauge the performance of the lenses. As the practical distance at examination has been reported to be an important factor, we used the distances that are within arm’s reach when the patient holds a test chart; because Japanese people have relatively short average height and arm length compared with European and American people, we measured intermediate visual acuity at 50 cm, which was assumed to be within the reach of most patients in this study.
Another limitation is that potential differences in the socioeconomic status might exist between the patients in the two groups in this study. To increase the validity of the study, this large-scale, single-center study was performed under a consistent protocol as described in our previous study11,12, i.e., after written informed consent was obtained from all the patients before surgery, we evaluated the same series of pre- and postoperative parameters, including the VFQ-25 score. Although this study was retrospective, each patient who was implanted with lenses was randomly and independently sampled, and all endpoints were measured. Furthermore, we strictly adjusted for age, sex, axial length, subjective refraction SE, subjective refraction CYL, corneal astigmatism (keratometric cylinder), corneal higher-order aberrations, and pupil diameter in the same way described in our previous study11,12. Because the data contain a mixture of items evaluated either in both eyes or in each eye separately, we implemented an analysis that accounted for bias by using a linear mixed model and corrected for multiple observations for each eye per patient. It is common in statistics to assume that random assignment does not bias the results of the analysis, even if there are differences in the number of patients, such as the 1:n allocation used in clinical trials. In all multivariate analyses, the missing values of explanatory variables were filled in by the multiple imputation method with 10 imputations.
In conclusion, we compared the comprehensive visual performance of a diffractive trifocal IOL with a diffractive bifocal IOL, which is another multifocal lens of a similar diffractive type. Patients in the bifocal group had better uncorrected near visual acuity, whereas patients in the trifocal group had better uncorrected intermediate visual acuity and contrast sensitivity (with/without glare). At high-performance levels, the two IOL groups had different characteristics regarding various visual parameters.