Approximately 47% of patients have at least 0.75 D of astigmatism in at least one eye,1 but astigmatism correcting lenses (TCLs) are only fitted on approximately 25% of soft contact lens (CL) wearers.2,3 Without astigmatism correction, subjects have reduced high- and low-contrast visual acuity at distance and near, reduced contrast sensitivity, and more difficulty with everyday tasks such as reading, computer work, and driving.4 Correcting astigmatism has been shown to improve these objective measures and patient satisfaction.4 A study by Milton et al.5 fit habitual CL wearers with spherical equivalent, daily disposable CLs and used specially designed trial lenses over the spherical equivalent CLs to show subjects the difference in vision with and without toric correction. This study reported that 91.9% of subjects preferred toric correction when looking at scenes, faces, and other “real world” visual stimuli.
Previous studies have shown that high- and low-contrast visual acuity are improved with TCLs compared with spherical contact lenses (SCLs) in astigmats, which suggest that TCLs are beneficial for patients.6–9 However, it is important to consider whether patients can subjectively appreciate these improvements in their day-to-day lives beyond visual outcomes. This study aimed to assess if subjective, patient-reported measures are improved with soft TCL compared with soft SCL in patients with mild-to-moderate astigmatism and if clinician time needed to fit the TCL is significantly greater compared with SCL.
This study was approved by the University of Houston Committee for the Protection of Human Subjects, and complied with the tenets of the Declaration of Helsinki. After providing informed consent for participation, subjects provided a medical and CL wearing history. Subject eligibility included the following: (1) between 18 and 45 years of age, inclusive; (2) no need for presbyopic correction; (3) habitual soft CL wearer with a reported average wear frequency of at least 5 days a week and at least 8 hrs a day in the month before the initial visit; (4) spherical refraction at the corneal plane between +4.00 and +0.25 D or −0.50 and −9.00 D, inclusive; and (5) cylinder power at the corneal plane between −0.75 and −1.75 DC, inclusive. Subjects with eye abnormalities beyond refractive correction (e.g., previous ocular surgeries, slitlamp examination findings that would contraindicate CL wear, active ocular infection, and strabismus) were not eligible for the study.
In this cross-over clinical trial, all eligible subjects were fitted binocularly with TCL (1-Day ACUVUE MOIST for Astigmatism; Johnson & Johnson Vision Care, Inc, Jacksonville, FL) and a SCL (1-Day ACUVUE MOIST) one at a time during two separate wear periods. At each fitting visit (study visit 1 and visit 3), subjects were fitted into either the TCL or SCL based on a block randomization sequence. For each CL type, the number of CL changes needed to achieve a successful CL fit was recorded. At the fitting visit, a time stamp was used to document the beginning and successful completion of the CL fitting process. Trained examiners following a standardized fitting protocol inserted and removed all CLs to avoid inter-subject differences in CL insertion techniques and habits. A successful CL fit was defined as achieving an over-refraction of plano with each eye. Overlabels were used to mask subjects to the CL type (TCL or SCL) throughout the study.
During each wear period (between visits 1 and 2, then visits 3 and 4), subjects were instructed to wear the study CLs for at least 5 hrs per day for at least 5 days. The follow-up visit was scheduled 5 to 10 days after fitting visit, and subjects were instructed to have worn the study CLs continuously for at least 2 hrs before the start of each visit to ensure CL stability and adequate time for adaptation. If the subject was wearing TCLs, the number of degrees of rotation in primary gaze were documented. Clinicians also documented the number of degrees of rotation from the primary gaze position during a series of unforced blinks. Subjects were asked to look in one position of gaze (up, down, left, or right), blink, and then return to the primary position of gaze. This was repeated for each of the positions of gaze. Clinicians documented the largest amount of rotational deviation from primary gaze after looking in a cardinal direction.
At baseline and at each follow-up visit, subjects completed a National Eye Institute Refractive Error Quality of Life Instrument (NEI-RQL-42)10 and the Convergence Insufficiency Symptom Survey (CISS).11,12 Modifications to the NEI-RQL-42 included: (1) the replacement of the phrase “…glasses, contact lenses, magnifier, or other type of correction (including surgery) you have?” with “…correction you are wearing now” in two questions; and (2) the replacement of “in the last 4 weeks” with “in the last week” in several questions to ensure that subjects answered based on the study period when they were wearing their study CLs. There was a 5- to 10-day wash-out period between wear periods (visit 2 and visit 3) in which subjects wore their habitual CL or spectacles. Each follow-up visit was conducted identically with the exception that subjects were asked if they preferred one CL over the other (first lens used vs. second lens used) at visit 4, and, if a preference was indicated, to identify which set of lenses were preferred.
The sample size for this study was calculated before the initiation of data collection. The outcome measures were identified as the differences in the NEI-RQL instrument overall scale, clarity of vision subscale, and diurnal fluctuation subscale scores using a paired t test comparing the difference between conditions (TCL vs. SCL) to zero. For each scale, a 10% difference between conditions was considered as the effect size with a power of 90% and 2-sided type I error of 0.01. The previously reported SD for the difference in each scale between two administrations of the NEI-RQL are clarity of vision (11.4), diurnal fluctuations (13.9), and overall (4.9). Assuming an intraclass correlation of 0.50, the scale associated with the largest sample size was the diurnal fluctuation scale with a calculated sample size of 39. To account for potential loss to follow-up or missing data points in the final data set, the final sample size of the study was set to 60 CL wearing subjects.
A generalized linear model with negative binomial distribution was used to evaluate the number of CLs needed to achieve a successful fit with the TCL versus SCL. This model controlled for sequence of CL wear, CL type, and wear period.
A linear mixed model was used to evaluate the successful time on lens insertion (time-to-fit) with the TCL and SCL. Along with CL type, fixed effects in this model included a sequence of CL wear and wear period.
For the NEI-RQL-42, both global score and subscale scores were calculated, and the overall CISS score was calculated. For the NEI-RQL-42, higher scores are better. For the CISS, lower scores are better (indicate fewer symptoms). Because the normality assumption was not met for all NEI-RQL-42 subscale scores, the scores were compared between TCL and SCL using nonparametric Wilcoxon tests. The P-values were corrected to account for multiple comparisons using the Benjamin–Hochberg method.13,14 A linear mixed model, which controlled for sequence of CL wear, CL type and wear period, was used to compare CISS scores while wearing the TCL versus SCL.
Sixty subjects (71.7% women) with a mean age (±SD) of 27.5±5.0 years completed the study. The spherical mean (±SD) refractive error was −3.68±2.01 D, and the cylindrical mean was −1.28±0.36 D.
The average number of CLs (±SD) needed to fit the TCL was 1.2±0.47, whereas it was 1.2±0.46 for the SCL (least square [LS] mean ratio [TCL/SCL] = 0.9; 95% CI=0.8–1.1; P=0.68). Rates of having a successful fit with the initial lens chosen by examiners were 84.2% for TCL and 79.2% for SCL. It took an average of 10.2±4.3 min to fit the TCL, and 9.0±6.5 min to fit the SCL ([toric−sphere] = 1.2; 95% CI=−1.1–3.5; P=0.22). In primary gaze, the mean amount of rotation for the TCL was found to be 4.6°±5.8°. On blinking, the mean rotation for the TCL was 0.9°±2.4°, and with gaze movements, the mean rotation was 2.8°±3.5°.
Table 1 shows NEI-RQL-42 and CISS scores with the TCL and SCL as well as the P-values resulting from comparisons between scores when wearing TCL compared with SCL at follow-up. Toric contact lens scored significantly better (higher values are better) than SCL in global NEI-RQL-42 score (P=0.006) and the clarity of vision (P=0.006) and satisfaction with correction subscales (P=0.006). The TCL also scored better (lower values are better) than the SCL on the CISS. The linear model for CISS provided an LS mean difference (TCL−SCL) of −2.20 (95% confidence interval: −4.25 to −0.14) showing a 15% reduction in symptoms with TCL compared with SCL (P=0.02).
Fifty-five subjects (91.7%) reported having a CL preference. Those who did not report a preference were either fitted binocularly with −0.75 DC CL (n=3) or with a −0.75 DC CL in one eye and a −1.25 DC CL in the other (n=2). A total of 43 subjects (72%) preferred the TCL compared with the SCL (binocularly fit with −0.75 DC: n=10; fit with −0.75 DC in one eye and −1.25 in the other eye: n=16; binocularly fit with −1.25 DC: n=17). Those who preferred the SCL included six subjects who were fitted binocularly with −0.75 DC CL, one subject who was fitted binocularly with −1.25 DC CL, and five who were fitted a −0.75 DC CL in one eye and a −1.25 DC CL in the other.
This study shows that there was no significant difference between TCL and SCL in time to successful CL fit or the number of CLs used in the fitting. Most subjects also preferred TCL to SCL. Although every effort was made to mask subjects to the CL type (TCL vs. SCL), the authors acknowledge that some astute subjects may have noted differences with fitting procedures or with differences in lens design, such as the lens mark designed to assess lens rotation in the TCL. However, it is unlikely that these observations contributed significantly to the preference for TCL indicated by the subject responses.
To our knowledge, reports that compared fitting times and number of lenses needed to successfully fit a TCL and SCL are not present in the literature. A study by Morgan et al.6 did not specifically look at ease of TCL fitting, but did report that 90% of TCL fits were successful with the first CL, and 66.7% of those that did not successfully fit with the first CL were successfully fit with the second CL. The ease in fitting reported by Morgan et al. supports the findings of this study regarding the fitting of TCL.
This study also showed that the lenses were very stable in primary gaze, with blink, and with eye movement. This stability may contribute to the improvements in subjective vision with TCL as well as preference for TCL. A previous study by Zikos et al. reported on lens rotation with eye versions in two different lenses using an infrared head-mounted video system.15 The lens tested in the current study provided a similar amount of rotation compared with one lens tested in the Zikos et al. study (3.39°±6.59°) and a reduced amount compared with the other lens (−8.08°±12.93°—note that the negative value indicated a counterclockwise rotation and a plus value indicated a clockwise rotation). A previous study did not find evidence to support the ability of TCL stability to predict visual performance; however, this abstract did not report on subjective data.16
The improvements in subjective vision reported here are consistent with previous reports. Sulley et al.17 reported that 90% of subjects who habitually wore SCL ranked their vision as “excellent,’ “very good,” or “good”; however, this percentage increased to 98% after the sample was fit with a TCL. Although the Sulley et al. report agrees with the findings shown here, a study by Gaib and Vasudevan9 did not find subjective improvements in vision with custom TCL compared with SCL. Differences in methods, in addition to differences in lens design, could have contributed to the difference in results between the current study and the study by Gaib and Vasudevan. First, Gaib and Vasudevan only required that one eye need astigmatism correction, instead of both eyes as in this study. In addition, the subjects were only allowed to wear the lenses for 1 day (6 hrs) before providing their opinion in vision. In addition, the subject demographics are different between the Gaib and Vasudevan study and the present study. Sixty percent of subjects only had −0.50 D of refractive astigmatism in the Gaib and Vasudevan9 study with the maximum refractive astigmatism being −1.00 DC, whereas the present study's sample included subjects with low-to-moderate amounts of astigmatism (up to −1.75 DC). Another difference between studies by both Sulley et al. and Gaib and Vasudevan is the method of gathering the subjective data. The present study used two questionnaires, which probed many aspects of vision. In the previous studies, overall rankings were used, which makes the information gathered here unique and potentially valuable for eye care practitioners when prescribing for patients who may complain of clarity of vision or lack of satisfaction with SCL.
Preference data can be more easily compared across studies because of similarity in question format. A study by Cho et al. asked subjects whether they preferred their vision with a TCL or SCL. They reported that 58.5% preferred the TCL, 22.0% had no preference, and 19.5% reported preferring vision with the SCL.8 In the present study, the percentage of subjects who preferred TCL was higher (72%). This difference may be attributed to the fact that Cho et al. did not dispense CLs to the subjects for out-of-office wear. Therefore, subjects may not have been able to experience the benefits of TCL in their everyday lives. The difference could also be attributed to the toric lens design tested in the current study and the stability it showed in the eye. It is important to note that the current distribution suggests that subjects with more moderate astigmatism prefer TCL compared with SCL.
It is important to remember that only one type of TCL and one type of SCL were evaluated in this study. This consistency allowed for the CL material, method of stabilization, base curve, CL diameter, and edge design to be maintained across subjects and, thus, minimized the influence of comfort, which could confound the study results related to vision. It also eliminated any influence of CL solutions because the daily disposable modality was chosen. However, practitioners should consider this study's results in combination with the knowledge that differences in TCL design and stabilization method could influence visual acuity, higher-order aberrations, and comfort.7,18
Overall, TCL provided better subject-reported vision. Most subjects chose a TCL over a SCL. In addition, the improved visual function found when fitting TCL can be achieved with similar effort from the practitioner in time for fitting or the number of CLs needed to achieve a successful fit.
The authors like to acknowledge the efforts of the entire study team. A portion of this work was presented at the American Academy of Optometry 2015 Annual Meeting, Global Specialty Lens Symposium 2016, and the British Contact Lens Association and Netherlands Contact Lens Congress 2016. Clinicaltrials.gov Registration: NCT01857102.
1. Young G, Sulley A, Hunt C. Prevalence of astigmatism in relation to soft contact lens fitting. Eye Contact Lens 2011;37:20–25.
2. Nichols JJ. Contact lenses 2015. Contact Lens Spectr 2016;31:18–23, 55.
3. Morgan P, Woods C, Tranoudis I, et al. International contact lens prescribing in 2015. Contact Lens Spectr 2016;31:24–29.
4. Read SA, Vincent SJ, Collins MJ. The visual and functional impacts of astigmatism and its clinical management. Ophthalmic Physiol Opt 2014;34:267–294.
5. Milton A, Murphy M, Rose B, et al. Demonstrating correction of low levels of astigmatism with realistic scenes. Cont Lens Anterior Eye 2016;39:26–37.
6. Morgan PB, Efron SE, Efron N, et al. Inefficacy of aspheric soft contact lenses for the correction of low levels of astigmatism. Optom Vis Sci 2005;82:823–828.
7. Richdale K, Berntsen DA, Mack CJ, et al. Visual acuity with spherical and toric
soft contact lenses in low- to moderate-astigmatic eyes. Optom Vis Sci 2007;84:969–975.
8. Cho P, Cheung SW, Charm J. Visual outcome of soflens daily disposable and soflens daily disposable for astigmatism in subjects with low astigmatism. Clin Exp Optom 2012;95:43–47.
9. Gaib SN, Vasudevan B. Low toric
soft contact lens acceptance study. Cont Lens Anterior Eye 2015;38:194–198.
10. Nichols JJ, Mitchell GL, Saracino M, et al. Reliability and validity of refractive error-specific quality-of-life instruments. Arch Ophthalmol 2003;121:1289–1296.
11. Rouse MW, Borsting EJ, Mitchell GL, et al. Validity and reliability of the revised convergence insufficiency symptom survey in adults. Ophthalmic Physiol Opt 2004;24:384–390.
12. Rouse M, Borsting E, Mitchell GL, et al. Validity of the convergence insufficiency symptom survey: A confirmatory study. Optom Vis Sci 2009;86:357–363.
13. Benjamini Y, Hochberg Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc Ser B (Methodological) 1995;57:289–300.
14. Ferreira J, Zwinderman A. On the Benjamini–Hochberg method. Ann Stat 2006;34:1827–1849.
15. Zikos GA, Kang SS, Ciuffreda KJ, et al. Rotational stability of toric
soft contact lenses during natural viewing conditions. Optom Vis Sci 2007;84:1039–1045.
16. Chamberlain P, Morgan PB, Moody KJ, et al. Fluctuation in visual acuity during soft toric
contact lens wear. Optom Vis Sci 2011;88:E534–E538.
17. Sulley A, Young G, Lorenz KO, et al. Clinical evaluation of fitting toric
soft contact lenses to current non-users. Ophthalmic Physiol Opt 2013;33:94–103.
18. Berntsen DA, Merchea MM, Richdale K, et al. Higher-order aberrations when wearing sphere and toric
soft contact lenses. Optom Vis Sci 2009;86:115–122.