Purpose
To evaluate the effect of the Boston Ocular Surface Prosthesis (Boston Foundation for Sight) on higher-order wavefront aberrations in eyes with keratoconus, eyes that have undergone penetrating keratoplasty, eyes that have undergone refractive surgery, and eyes with ocular surface diseases.
Design
Prospective, clinical study.
Methods
The study evaluated 56 eyes of 39 patients with irregular astigmatism who were treated with the Boston Ocular Surface Prosthesis when conventional treatments failed. Patients were sorted into 4 clinical groups based on the underlying cause of irregular astigmatism, including keratoconus (group 1), post–penetrating keratoplasty (group 2), post–refractive surgery (group 3), and ocular surface diseases (group 4). Another 6 eyes of 5 patients who were treated with rigid gas permeable lenses also were evaluated. Best-corrected visual acuity; topographic refractive indices, including spherical, cylindrical, spherical equivalent values; and higher-order and total wavefront aberration errors were noted at baseline and after fitting the lens.
Results
In all groups, higher-order wavefront aberration error was noted to decrease significantly in eyes wearing the Boston Ocular Surface Prosthesis ( P < .001, P = .001, P = .002, and P = .001, respectively). By post hoc analysis, significant differences in the level of higher-order aberrations were observed only between groups 1 and 4 ( P = .012) and groups 1 and 2 ( P = .033). In the overall group, mean correction rate of higher-order aberration error with the Boston Ocular Surface Prosthesis was 72.3%. However, in eyes with rigid gas permeable lenses, 2 eyes demonstrated increased higher-order aberration error, whereas the mean correction rate in other 4 eyes was only 42.5%.
Conclusions
With its unique structure, the Boston Ocular Surface Prosthesis was found to be very effective in reducing higher-order wavefront aberrations in patients with irregular astigmatism resulting from a number of corneal and ocular surface conditions who had not responded satisfactorily to conventional methods of optical correction.
Irregular astigmatism and higher-order aberrations resulting from severe keratoconus are common causes of reduced quality of vision in most patients with that disease and in certain patients with a history of complicated refractive surgery, penetrating keratoplasty (PK), or ocular surface disease. Visual rehabilitation of patients with irregular astigmatism is challenging because conventional methods of optical correction often do not provide satisfactory quality of vision.
Based on the severity of refractive errors and underlying causes, spectacles and soft and rigid gas permeable contact lenses may provide visual rehabilitation in some of the cases. Recently, some invasive procedures such as intracorneal ring segment implantation and corneal collagen cross-linking alone or in combination have been reported to improve vision in these challenging cases. However, these therapies are not effective in all cases, and their effects on reducing higher-order aberrations remains to be determined.
The Boston Ocular Surface Prosthesis (Boston Foundation for Sight, Needham, Massachusetts, USA) is a specially designed gas-permeable scleral lens that was approved by the Food and Drug Administration in 1994. Each device is manufactured using computer-aided design and a computerized lathe to obtain strict tolerances and a precise fit. It has been used to improve visual function in corneal diseases with high irregular astigmatism, such as severe keratoconus that cannot be corrected adequately with spectacles or other types of contact lenses.
The aim of this prospective clinical study was to evaluate the impact of the Boston Ocular Surface Prosthesis on refractive errors and higher-order wavefront aberrations in patients with irregular astigmatism resulting from a variety of conditions that include keratoconus, PK, refractive surgery, and ocular surface diseases.
Methods
Study Population
The study included 56 eyes of 39 patients (17 female and 22 male) who were referred for treatment of reduced visual function resulting from irregular astigmatism and who did not have satisfactory improvement in vision from conventional treatments that included spectacles and soft or rigid gas permeable (RGP) corneal contact lenses. Eyes were stratified into 4 groups based on the cause of irregular astigmatism. The groups included 18 eyes with keratoconus (group 1), 10 eyes with a history of PK (group 2), 10 eyes with a history of refractive corneal surgery (3 radial keratotomy, 2 radial keratotomy and laser in situ keratomileusis (LASIK), 4 corneal rings, 1 LASIK; group 3), and 18 eyes with a variety of cornea and ocular surface diseases, such as Sjögren syndrome, neurotrophic corneal scarring, postherpetic corneal scarring, and dysfunctional tear syndrome (group 4). For further analysis, 6 eyes of 5 patients wearing RGP corneal contact lenses also were evaluated for study parameters. RGP corneal lenses included spherical, keratoconus, and postgraft designs.
Fitting of the Boston Ocular Surface Prosthesis and training on its use was performed by the same person (A.G.). During the consultation, a trial lens was inserted with and without front surface eccentricity. If eccentricity was determined to improve vision, then varying amounts of eccentricity were attempted (0.3, 0.6, and 0.8) until there was no further improvement in measured visual acuity or the patient did not report an appreciable difference in quality of vision. Of the 56 Boston Ocular Surface Prostheses fit, eccentricity was added to 16 devices.
Visual Acuity
Best-corrected visual acuity (BCVA) was measured under standardized room and chart illumination with the best manifest spectacle refraction or previously worn contact lens at baseline and with the Boston Ocular Surface Prostheses using the Marco Epic 5100 (Marco, Jacksonville, Florida, USA). BCVA values were converted to logarithm of the minimal angle of resolution values for statistical analysis.
Higher-Order Aberrations
Higher-order wavefront aberrations were measured with the NIDEK OPD-Scan II (Optical Path Difference Scanning System ARK-1000, software version 2.10c; Nidek Co, Ltd., Jacksonville, Florida, USA) before and while wearing the Boston Ocular Surface Prosthesis after fitting. All measurements were performed by the same examiner (A.G.) with the optical measurement zone set at 4 mm and the radial order set at 4. Three images were recorded for each eye across a natural (undilated) pupil. In addition to spherical and cylindrical refractive errors, both total and higher-order aberration error values were recorded in each measurement.
The OPD-Scan optical path scanning system generates a wavefront higher-order aberration map that displays specific higher-order aberration components, extracted from the total wavefront map. Moreover, this map illustrates the location and degree of higher-order aberrations in the eye.
Statistical Analysis
SPSS software version 15.0 for Windows evaluation version (LEAD Technologies, Inc, Chicago, Illinois, USA) was used for statistical analysis. Differences between the groups in terms of age and sex distribution were analyzed using a 1-way analysis of variance and chi-square tests, respectively. Paired-samples t tests were performed to evaluate the differences in study parameters before and after wearing the Boston Ocular Surface Prosthesis. A repeated-measures of analysis of variance was carried out to compare the difference in study parameters at baseline and while wearing the Boston Ocular Surface Prosthesis between the groups. A P value ≤ .05 was considered to be statistically significant.
Results
In the overall patient group, the mean age of 39 patients was 46.9 years, ranging from 23 to 75 years. There was no statistically significant difference between the 4 groups in terms of age and sex distribution ( P > .05).
When the patients were analyzed as a whole, statistically significant differences in all study parameters, including logarithmic BCVA, total and higher-order wavefront errors, spherical errors, cylindrical refractive errors, and spherical equivalent were noted before and after wearing the Boston Ocular Surface Prosthesis.
Results of study parameters in 4 groups at baseline and while wearing the Boston Ocular Surface Prosthesis are provided in Table 1 . BCVA increased significantly in all groups with the Boston Ocular Surface Prostheses (< 0.001, 0.001, 0.006, 0.001, respectively). The mean improvement in BCVA was at least 3 lines. There was a statistically significant reduction in spherical error in eyes wearing the Boston Ocular Surface Prosthesis in all groups (< 0.001, < 0.001, 0.035, respectively) except for group 4 ( P = .229). A significant reduction in cylindrical errors was observed in all groups (all P ≤ .001). Only groups 1 and 2 had a significant change ( P < .001) in spherical equivalent.
| Group | Parameter | Baseline | With the BOSP | Change | P Value |
|---|---|---|---|---|---|
| Group 1 (n = 18) | BCVA | 0.59 ± 0.46 | 0.09 ± 0.10 | + 4 lines (1 to 10) | <.001 |
| ZS error | −20.2 ± 12.2 | −0.25 ± 0.55 | −19.9 ± 12.2 | <.001 | |
| ZC error | 7.94 ± 6.15 | 0.97 ± 0.88 | 6.97 ± 6.26 | <.001 | |
| SE | −16.2 ± 10.8 | 0.24 ± 0.64 | −16.5 ± 11.1 | <.001 | |
| HO error | 1.89 ± 0.83 | 0.38 ± 0.29 | 77.1% (32.0 to 96.2) | <.001 | |
| Total error | 7.97 ± 3.59 | 0.98 ± 0.64 | 86.8% (75.0 to 95.3) | <.001 | |
| Group 2 (n = 10) | BCVA | 0.76 ± 0.29 | 0.14 ± 0.17 | + 5 lines (0 to 8) | .001 |
| ZS error | −12.6 ± 6.10 | −0.47 ± 0.70 | −12.2 ± 5.74 | <.001 | |
| ZC error | 7.87 ± 3.09 | 0.70 ± 0.19 | 7.17 ± 3.08 | <.001 | |
| SE | −8.71 ± 5.40 | −0.12 ± 0.66 | −8.59 ± 5.08 | <.001 | |
| HO error | 1.06 ± 0.59 | 0.25 ± 0.16 | 72.7% (41.9 to 91.5) | .001 | |
| Total error | 5.21 ± 1.80 | 0.65 ± 0.23 | 86.8% (78.2 to 93.3) | <.001 | |
| Group 3 (n = 10) | BCVA | 0.48 ± 0.40 | 0.04 ± 0.08 | + 4 lines (0 to 10) | .006 |
| ZS error | −5.65 ± 6.22 | −0.50 ± 0.72 | −5.15 ± 6.55 | .035 | |
| ZC error | 3.40 ± 1.81 | 0.45 ± 0.40 | 2.95 ± 1.77 | .001 | |
| SE | −3.95 ± 5.45 | −0.27 ± 0.73 | −3.67 ± 5.86 | .079 | |
| HO error | 1.32 ± 0.71 | 0.33 ± 0.24 | 69.2% (16.3 to 93.5) | .002 | |
| Total error | 3.47 ± 1.98 | 0.68 ± 0.33 | 74.9% (45.2 to 92.2) | .002 | |
| Group 4 (n = 18) | BCVA | 0.52 ± 0.51 | 0.14 ± 0.16 | + 3 lines (0 to 7) | .001 |
| ZS error | −1.87 ± 4.36 | −0.63 ± 1.03 | −1.24 ± 4.20 | .229 | |
| ZC error | 3.08 ± 2.73 | 0.58 ± 0.40 | 2.50 ± 2.57 | .001 | |
| SE | −0.33 ± 3.76 | −0.34 ± 0.99 | 0.01 ± 3.64 | .987 | |
| HO error | 0.98 ± 0.84 | 0.22 ± 0.14 | 70.1% (42.9 to 89.8) | .001 | |
| Total error | 2.98 ± 2.18 | 0.64 ± 0.39 | 69.2% (35.1 to 96.8) | <.001 |
The graphs showing which types of higher-order aberrations were predominant at baseline and while wearing the Boston Ocular Surface Prosthesis in both groups are provided in Figures 1 and 2 . Mean values of trefoil and coma were more than 0.5 μm in the keratoconus group ( P > .05), and trefoil was above this threshold in the post-PK and post–refractive surgery groups. At baseline, the higher-order aberrations accounted for 32.4% of the total aberration in the entire group. In the 4 different disease groups, these values were 25.1%, 20.7%, 44.1%, and 39.6%, respectively.
The change in higher-order error values is provided in Figure 3 . The ratio of higher-order error reduction that could be achieved by the Boston Ocular Surface Prosthesis wearing was the highest in keratoconus eyes, as much as 77.1%. Details of both higher-order and total aberration error reduction are provided in Table 1 .
In a separate analysis, we evaluated whether there was any difference in the magnitude change in wavefront aberrations with the Boston Ocular Surface Prosthesis among the 4 different disease groups. Based on the post hoc analysis, there were statistically significant differences in terms of higher-order error only between groups 1 and 4 ( P = .012) and between groups 1 and 2 ( P = .033). For total error, the difference reached statistical significance between group 1 and groups 2, 3, and 4 (group 1 vs group 2, P = .041; group 1 vs group 3, P = .001; and group 1 vs group 4, P < .001).
Another analysis was performed to determine if there were differences between eyes fit with lenses with eccentricity compared with those without. No statistically significant difference was found in BCVA ( P = .745), total ( P = .176), or higher-order aberration error values ( P = .993) between these 2 groups. However, a significantly higher reduction in total spherical aberration was observed in the group with eccentricity (correction, 0.32 ± 0.36) compared with those without eccentricity (correction, 0.12 ± 0.33; P = .05; Figure 4 ).
