Abstract
purpose. The purpose of this study was to measure upper and lower tear menisci on contact lenses using real time optical coherence tomography (OCT).
methods. Both eyes of 20 adapted contact lens wearers were imaged when they wore two types of silicone hydrogel lenses, balafilcon A on one eye and galyfilcon A on the other eye. The height, radius, and area of upper and lower tear menisci were obtained before, immediately after, and 20 minutes after lens wear on two consecutive days. On a third visit, the lenses were switched between eyes and an investigative lubricant was instilled after 4 hours of lens wear. OCT images were obtained as before and additionally at 1 and 4 hours after lens insertion. Imaging was also conducted immediately after lubricant instillation and 20 minutes post-instillation.
results. There were no significant differences in the measured variables between lenses (P > 0.05) and between consecutive visits (P > 0.05). The menisci around both eyelids immediately after lens insertion were significantly greater than that before (P < 0.005). By 20 minutes after lens insertion, all variables had returned to baseline values and remained that way for at least 4 hours. Immediately after the instillation of the lubricant at 4 hours, meniscus variables increased (P < 0.001), but recovery to baseline occurred within 20 minutes.
conclusions. It is feasible to use OCT in the measurement of both upper and lower tear menisci on contact lenses in situ. Tear menisci increased on insertion of contact lenses and on instillation of lubricants. However baseline values were re-established with 20 minutes.
Soft contact lenses must remain wet during lens wear so that sharp vision and ocular comfort are maintained. It is very common for soft contact lens wearers to experience dry eye, especially in the afternoon after wearing them for a period of time.
1 2 Dry eye is a common cause for wearers to abandon use of their lenses.
1 2 Changes in tear distribution and dynamics during lens wear remain unknown due to the extreme difficulty of imaging both tear menisci at the same time. Thanks to the advancement of real-time imaging and recording with optical coherence tomography (OCT), tear dynamics were successfully imaged recently.
3 Here we used OCT to simultaneously image the tear menisci between contact lenses and the upper and lower eyelids.
Ethics clearance was obtained from the research review board of the University of Rochester. Informed consent was obtained from all participants before enrollment in the study in accordance with the tenets of the Declaration of Helsinki. A total of 20 participants (5 men and 15 women; mean ± SD age: 31.7 ± 8.9 years) were recruited. All subjects were experienced contact lens wearers and were wearing soft contact lenses before participation in this study. As described previously,
3 the study was conducted in a small consulting room with monitored temperature (15°C–25°C) and humidity (30%–50%). At each subject’s screening visit, the contact lens prescription of both eyes was obtained based on the existing prescription and auto-refraction.
Two silicone hydrogel lenses were used in this study. They were PureVision (balafilcon A; Bausch & Lomb, Rochester, NY) with base curve 8.6 mm and Acuvue Advance (galyfilcon A; Vistakon, Jacksonville, FL) with base curve 8.3 mm. The diameter of both lenses was 14 mm. Contact lens solution (Opti-Free RepleniSH; Alcon, Fort Worth, TX) was used to soak these study lenses for 8 hours before insertion by subjects for visits one and two. A different contact lens solution (ReNu Multiplus; Bausch & Lomb) was used to soak the lenses for visit three. Lenses from the same lens package were used on the same eye during the first two visits, and the lens from a new package was used for the third visit.
During OCT imaging, a vertical scan across the corneal apex was conducted while the subject fixated on an external target and blinked normally. During the first two study visits, the procedure was performed at the same time of day on consecutive days. To enable statistical tests for repeatability of measurements, the PureVision lens was always inserted in the right eye and the Acuvue lens was always inserted in the left eye of each subject on both visits. The choice for right and left eyes was made randomly at the beginning of the study. OCT images
(Fig. 1)were obtained at three checkpoints: before, immediately after, and 20 minutes after lens insertion.
During the third visit, the lenses were switched between eyes. After wearing the lenses for 4 hours, 35 μL of a contact lens rewetting drop (Bausch & Lomb) was instilled into the eye when the lens was in situ. OCT imaging during visit three was conducted before lens wear, immediately after lens insertion, 20 minutes, and 1 and 4 hours after lens insertion. OCT images were also taken immediately after the instillation of the lubricant and 20 minutes after the instillation while the lens was still on the eye.
The OCT configurations for imaging the tear menisci were reported previously.
3 Briefly, a real-time OCT instrument, using a light source with a center wavelength of 1310 nm and a 60 nm bandwidth, was connected to a telecentric probe. The probe was mounted with a standard slit-lamp. Parallel beam scanning enabled the capture of spatial information for the tear menisci between the lens and the eyelids. The scan width was 12 mm and speed was 8frames per second. A video viewing system was coaxially aligned with the OCT scan. Alignment was based on observation of a specular reflex from the corneal apex.
Custom software was used to process the OCT images to yield all variables. Tear meniscus parameters were obtained from processing the first good image showing both upper and lower tear menisci taken between 1 and 2 seconds after a blink. The height of the tear meniscus was calculated from the image as the transverse distance. Pink dots identifying the front surface central point of the lens and the junction points between lens and upper and lower tear menisci were marked on the image
(Figs. 2A and 2B) . A circle was fitted to outline the lens surface using this three-point method. Then touch points between the tear meniscus and eyelid and between the tear meniscus and lens were marked with yellow dots, and the middle point of the tear meniscus front edge was marked with a green dot
(Figs. 2C and 2D) . After that, custom software processed the image to yield the results. The three-point method was used to fit a circle to yield the tear meniscus radius. The meniscus results included the upper tear meniscus radius (UTMR), height (UTMH), and area (UTMA), and the lower tear meniscus radius (LTMR), height (LTMH), and area (LTMA).
Values were presented as means ± standard deviations for all 20 subjects for visits one and two. However, due to image data corruption for one eye of one subject (ID 18), only 19 subjects were analyzed for the third visit. Data analysis was performed using statistical analysis software (Statistica, StatSoft, Tulsa, OK). Repeated ANOVA (re-ANOVA) was used to test significant differences of the measured means over time between visits, lenses, and solutions. The averaged differences of all measured variables obtained in visits one and two were also calculated. Pearson correlation was used to test correlations between variables. Repeatability of the measurements was calculated as the SD of the differences between two repeated measurements.
There were no significant differences among all measured tear menisci variables between types of lenses (re-ANOVA,
P > 0.05 for all six measures over the three checkpoints;
Table 1 ;
Fig. 1 ) or between visits one and two (
P > 0.05;
Table 1 ). The menisci immediately after lens insertion were significantly greater than those before lens wear (
P < 0.005;
Table 1 ;
Figs. 1 and 3 ). By 20 minutes after lens insertion, the values of tear menisci had recovered to the baseline values (
P > 0.05). The repeatability of menisci measurements at baseline, lens insertion, and 20 minutes after inserting the lens were unchanged (
P > 0.05 for all 6 measures over three checkpoints;
Table 2 ;
Fig. 3 ). Changes in UTMR, UTMA, and UTMH followed similar patterns to the changes in LTMR, LTMA, and LTMH immediately after insertion of the lenses and at 20 minutes later
(Table 1) . The changes of the upper and lower tear meniscus variables were highly correlated with one another immediately after lens insertion and 20 minutes later (Pearson correlation, r ranged from 0.45 to 0.94,
P < 0.05;
Fig. 4 ).
During the third visit, tear menisci around both upper and lower eyelids were significantly increased immediately after the instillation of the lubricant (
P < 0.001;
Fig. 5 ). By 20 minutes after instillation, all the variables had returned to baseline values (
P > 0.05). There were no significant differences between lenses for any measured variable of the tear menisci (
P > 0.05). Compared with results obtained in visits one and two, no differences were found between eyes and lens care solutions with each of tested lens (
P > 0.05).
The method of imaging dynamic tear distribution on the ocular surface using real-time corneal OCT was developed and validated previously.
3 4 The repeatability of the measurements was tested at baseline and after instillation of artificial tears.
3 4 This method is a non-contact, non-invasive, real-time modality. As such, it is suitable for imaging the tears of both menisci simultaneously during contact lens wear. Compared to previously reported methods in measuring tear menisci,
5 6 7 8 it has unique advantages in studying the tear system during contact lens wear. In this study, repeated measurements were conducted and no differences for each of the measured variables were found between repeated visits. The repeatability of the measurements at all test periods was comparable to the results determined in the eyes without contact lenses.
3 This confirms the good repeatability previously reported for this real-time OCT instrument,
3 4 which is also suitable to study contact lens wearers. It is capable of detecting changes in tear meniscus variables, as shown by the measurements recorded between the initial lens insertion and 20 minutes afterward, which is a normal settlement time for adapted contact lens wearers.
We tested two different types of lenses to determine whether differences in materials and/or manufacturers existed. However, there were no differences in the tear meniscus variables tested in the short term. Simultaneous measurements of both upper and lower tear menisci on contact lenses in situ have not been made previously, to the best of our knowledge. The reason might be the difficulty in imaging both eyelids at the same time with a slit-lamp or tear meniscometer.
9 Real-time imaging is necessary to capture the dynamic nature of the tear system and to define a specific time after a blink for taking the measurements. In this study, high quality images were obtained to process the dimension information at a specifically defined time, between 1 and 2 seconds, after a blink. Distorted images were often obtained during the first second after the blink. These distortions were possibly due to lens movement and lens deformation.
At lens insertion, increases in the tear menisci were evident. This may be due to extra fluid brought into the eye with the lenses and/or reflex tearing on insertion of the lens. The increased tear meniscus radius after lens insertion probably is related to the increase of the pre-lens tear film, as demonstrated by Nichols and King-Smith.
10 The relationship between tear meniscus radius and pre-corneal tear film thickness was also established by mathematical calculation.
11 After the 20-minute period of adaptation, tear menisci around both eyelids recovered to the baseline levels before lens wear. This suggests that wearing contact lenses might not alter the dimensions of tear menisci after a short time of adaptation.
Tear meniscus dimensions in non-contact lens wearers were studied previously using the same technique.
3 The baseline meniscus data are nearly identical with those found in the contact lens wearers of this study. This implies that the tear menisci remain unchanged when subjects wear properly fitted contact lenses, thereby confirming that the tear menisci remain unaltered during contact lens wear. Further studies with a longer term of lens wear may be needed with asymptomatic and symptomatic contact lens wearers, specifically those who have dry eye complaints.
The lubricant drop clearly increases the tear menisci, as evident in the third visit. However, the menisci return to the baseline within a short period. In the present study, the increase of the tear menisci recovered within 20 minutes. It is unclear whether the pre- and post-lens tear films increased after the instillation. Further studies may be needed to investigate this aspect. Interestingly, no further decreases of the tear meniscus were found after 4 hours of lens wear compared to the baseline and 20 minutes after lens wear. This may indicate that a dynamic balance was reached after the short period of adaptation. However, what happens after a long period of lens wear during the day, such as 10 hours, remains unknown.
We tested the repeatability of measurements in visits one and two, using the same lenses on the same eyes with one type of lens solution, and found no differences between eyes. In visit three, we switched lenses between eyes and used a different solution in an attempt to demonstrate differences between eyes and solutions. However, as with visits one and two, we did not find significant differences between eyes or solutions. In visit three, we also demonstrated the changes after 4 hours of lens wear and instillation of artificial drops. Since this is the first study investigating the changes of both tear menisci on contact lenses, comparisons between eyes (by switching lenses) and solutions provide some insights on further study designs.
There is not much in the literature regarding the differences in tear menisci on different contact lenses or the repeatability of the methods used to detect such differences. We did not calculate the sample size in this study for detecting possible differences between lenses and solutions. The method may not be sensitive enough with the sample size used in the present study to detect a difference if one exists. On the other hand, it may not be possible to know what difference might be clinically meaningful for the tear meniscus even if a difference had been found. Further studies will address this issue by investigating the relationship between tear menisci and ocular comfort.
In summary, we demonstrated for the first time that OCT is a useful tool in the measurement of tear menisci on contact lenses in situ. Differences of tear menisci were found only immediately after the insertion of two different types of lenses, and recovery to baseline occurred within 20 minutes. No differences in tear menisci were found between the tested lenses. Instillation of a lubricant increases tear menisci for only a short period. Further studies should be conducted to characterize tear dynamics after long term lens wear and the use of rewetting drops. The relationship between tear distribution and dry eye symptoms in contact lens wearers also warrants further investigation.
The authors thank Jayachandra R. Palakuru for assistance and Britt Bromberg of Xenofile Editing for providing editing services for this manuscript.