There is increasing evidence that scleral support rigid gas permeable contact lenses are suitable to compensate a wide range of corneal conditions derived from primary corneal disease and postsurgical complications and even in normal corneas [
The recent rebirth of SL fitting has been accompanied by a more predictable fitting process, but there is still a significant degree of uncertainty due to the few options of devices to objectively measure anatomical features of the ocular surface beyond the corneal area. Optical coherence tomography (OCT) and scleral topographers are some options that could have an important role during the fitting process; however, they are still not widely used in clinical practice all over the world [
The primary goal of the present study was to analyze the number of trial lenses and reorders required to obtain a satisfactory fitting and to evaluate the learning process from the clinician perspective by evaluating the changes in fitting over the time of enrollment. A secondary goal was to evaluate the differences in the fitting complexity between irregular and normal corneas.
This was a prospective dispensing, case series involving patients with primary corneal ectasia, penetrating keratoplasty, postsurgical ectasia, and regular corneas with high refractive errors between December 2015 and March 2017. The study was conducted at the Clinical and Experimental Optometry Research Lab (CEORLab), at University of Minho (Braga, Portugal). A total of 95 subjects were primarily recruited to participate in a study involving scleral supported contact lens fitting. Lenses were manufactured by Procornea (Eerbeek, Netherlands). Other relevant technical details of the contact lenses are presented in Table
Characteristics of the mini- and full-scleral lenses trial sets used in the present study.
Parameter | Mini-scleral lens | Full-scleral lens |
---|---|---|
Material | Boston XO |
Boston Equalens II |
|
||
Dk (ISO/Fatt) | 100 | 85 |
|
||
Central Thickness (-3.00 D) | 0.25 mm | 0.45 mm |
|
||
Diameter | From 15.20 to 18.00 mm in 0.40 mm steps | From 18.00 to 24.50 mm in 0.50 mm steps |
|
||
Back Optic Radius | 8.20 mm |
From 7.20 mm to 9.80 mm in 0.10 mm steps |
|
||
Power | Sphere +20.00 D to -25.00 D in steps of 0.25 D; Front cyl -0.50D to -3.00D in steps of 0.25D; Axis 0 to 180 degrees in steps of 1 degree | Sphere +30.00 D to -30.00 D in steps of 0.25 D; Front cyl -0.50D to -3.00D in steps of 0.25D; Axis 0 to 180 degrees in steps of 1 degree |
|
||
Refractive Index | 1.415 | 1.423 |
|
||
Hardness | 81/112 (Shore/ Rockwell) | 114 (Rockwell) |
|
||
Density | 1.27 | 1.24 |
|
||
Contact Angle (deg.) | 49 | 30 |
|
||
Sagittal height | From 0.25 to 6.75 in 0.25 steps | From 2.47 to 5.07 in 0.10 steps |
|
||
Peri-Factor / Sclera Opening | From -8 to +8 in steps of 1 | From 11.50 to 17.25 in 0.25 steps |
|
||
Toricity (difference in peri-factor) | From 1 to 6 in steps of 1 | From 1 to 4 in steps of 1 |
The subjects were divided into two major groups. One group (IC Group) comprised corneas with primary or secondary ectasias, postpenetrating keratoplasty, and other corneal irregularities due to refractive surgery or others. The second group comprised subjects with regular and healthy corneas (RC Group) that have failed or rejected other forms of vision correction with contact lenses, whether because of comfort or lens stabilization on eye (vision). Only subjects with moderate-to-high refractive errors (myopia > 6.00 D, astigmatism > 2.00 D, and/or hyperopia > 4.00 D) that failed other forms of vision correction were included in RC Group. Subjects with previous ocular surgery were excluded. Subjects of each group were further divided into subgroups for some analysis: Prim.IC included subjects from IC Group with primary ectasia or other conditions not induced by corneal surgeries and Sec.IC included those subjects from IC Group with secondary irregularities due to previous surgeries (corneal irregularities due to refractive surgery, penetrating keratoplasty, intracorneal ring segments implantation, and corneal cross-linking). Subjects from RC Group were separated according to their astigmatism into LA.RC (low astigmatism <2.00 D) and HA.RC (high astigmatism >2.00 D). To be included in the present study patients must have been dispensed with SL and have at least 1 follow-up visit completed (85 subjects).
Three repeated measures of corneal topography were done with Medmont E300 (Precision, Vancouver) in each eye in order to assess the severity of each case. Data from simulated keratometry (SimK), which measures the paracentral zone (usually 3 mm) of the anterior surface of the cornea and corneal asphericity (Q) of the flat and steep corneal meridians, were analyzed for each group. High and low contrast visual acuities (HCVA and LCVA, respectively) with ETDRS LogMAR scale charts were measured with habitual correction (HC) and best spectacle correction (BSC). Later, both HCVA and LCVA were also evaluated with SL.
All lenses were fitted by the same practitioner (R.M-A) who was a licensed optometrist with a Master Degree in advanced optometry but without previous clinical experience of scleral lens fitting. Prior to fitting the lenses, she received training on the fitting procedure. Following the recommendations of the declaration of Helsinki, all subjects received information from the study before they accept to participate and signed a consent form. The protocol of the study has been reviewed and approved by the Ethics Subcommittee for Life and Health Sciences of University of Minho.
All the subjects enrolled in this study had to attend several appointments during the follow-up: Baseline, lens dispensing visit, and follow-up visits: 1-, 3-, 6-, and 12-month visits. In this report, only subjects that were dispensed and have at least one follow-up visit were included. At the first appointment (Baseline) SL fitting was done. Fittings were performed using diagnostic fitting sets from Procornea. Subjects that were CL wearers previous to the trial visit were advised not to wear their habitual lenses 3 days before the Baseline appointment. The initial trial lens was determined following manufacturers’ guidelines, considering clinical features and the degree of severity of the corneal condition. Practitioner did not use any objective measurement that could aid in the selection of the first trial lens. All lenses were fitted empirically, based on trial and error process, with diagnostic lens sets. The best trial lens should align evenly on the scleral and vault the entire corneal surface and limbus, with a cornea-lens separation of about 300
When the ordered SLs arrived, subjects went to the lens dispensing visit (LDV), where the on-eye fittings were evaluated after lens insertion and after at least 90 minutes of lens wear. If the fitting was not satisfactory, another lens with different parameters was ordered (and was considered a reorder). The number of reorders at LDV (when needed) was recorded for each eye. Then, subjects were also evaluated at several follow-up appointments at 1, 3, and 6 months of lens wear (after LDV); reorders were also recorded at these visits. It was considered a “reorder” whenever it was necessary to order a new lens with different parameters for the same eye. Erroneous shipments and other factors not directly linked to practitioner fitting process were excluded from this analysis.
The number of trial lenses required to prescribe and order the lenses and the number of lenses reordered to the manufacturer at LDV and through the follow-up period were counted and grouped in 8 chronological groups in 20 fittings (eyes), without accounting for the group of the subject (IC or RC Group). Analysis involving the division into the different groups and subgroups was performed without chronological sequence.
Statistical analysis was conducted using SPSS version 24.0 (IBM Co, IL) to compare the number of trial lenses and reorders required between groups and subgroups. Normality of data distribution was analyzed with Shapiro Wilk test in different groups and subgroups. Pairwise comparison between groups or subgroups was done using an independent sample T-Test for normally distributed data and Wilcoxon signed ranks test for nonnormally distributed data. Multiple comparisons to evaluate the effect of time on number of trials and reorders or subject handling and wearing experience were evaluated with ANOVA test for normally distributed data and Kruskal-Wallis test for nonnormally distributed data. The level of statistical significance was set at p<0.05.
A total of 85 subjects (43 females and 42 males) with a mean age of 34.51±10.41 years were included in this report. Of them, 14 wore lenses in one eye and 71 wore lenses in both eyes representing a total of 156 eyes dispensed with SLs. Since not all fittings were bilateral, there were 5 cases in which both eyes of the same subject fell in different groups, which contributed to increase the chance of final homogenization between all groups. The fittings were divided into 8 groups of 20 fittings, in the chronological order of each fitting, in order to analyze the learning process. The sample was also analyzed separately according to the ocular condition that required the SL fitting in IC Group (irregular corneas, n=122 eyes) and RC Group (high refractive error, n=34 eyes).
Table
Demographic data of the patients analyzed in each clinical subgroup included in the present report.
Total | IC Group | RC Group | p | ||
---|---|---|---|---|---|
No. Subjects | 85 | 67 (79%) | 18 (21%) | - | |
|
|||||
No. Eyes Fitted | 156 | 122 (78%) | 34 (22%) | - | |
|
|||||
Gender | 43 female (51%) |
31 female (46%) |
12 female (67%) |
- | |
|
|||||
Subgroup (No. Fittings) | Prim.IC: 80 (66%) |
LA.RC: 8 (24%) |
- | ||
|
|||||
Age (years) | 34.51±10.41 |
35.54± 10.45 |
30.67±9.91 |
0.080+ | |
|
|||||
SimK Flat (D) | 43.93±5.51 | 44.20±6.19 |
42.99±1.62 |
| |
|
|||||
SimK Steep (D) | 47.29±6.04 | 47.78±6.74 |
45.58±1.71 |
| |
|
|||||
Q Flat | -0.65±0.53 | -0.71±0.58 |
-0.43±0.19 |
| |
|
|||||
Q Steep | -0.17±0.69 | -0.26±0.71 |
0.14±0.51 |
| |
|
|||||
HC |
Glasses | 73 | 45 | 28 | - |
Soft CL | 19 | 13 | 6 | - | |
RGP | 20 | 20 | 0 | - | |
Hybrid | 13 | 13 | 0 | - | |
SL | 11 | 11 | 0 | - | |
N/P | 20 | 20 | 0 | - | |
|
|||||
HCVA w/ HC | +0.30±0.30 | +0.34±0.31 |
+0.16±0.21 |
| |
|
|||||
LCVA w/ HC | +0.54±0.32 | +0.62±0.33 |
+0.31±0.18 |
| |
|
|||||
BSC | M (D) | -3.64±3.63 | -3.24±3.23 |
-4.94±4.57 |
0.078 |
J0 (D) | 0.23±1.02 | -0.04±0.92 |
1.09±0.89 |
| |
J45 (D) | 0.20±1.13 | 0.23±1.26 |
0.12±0.61 |
0.820 | |
|
|||||
HCVA w/ BSC |
+0.26±0.27 | +0.31±0.28 |
+0.11±0.17 |
| |
|
|||||
LCVA w/ BSC |
+0.51±0.30 | +0.58±0.29 |
+0.29±0.18 |
| |
|
|||||
HCVA w/ SL |
+0.07±0.15 | +0.08±0.15 |
+0.06±0.15 |
0.650+ | |
|
|||||
LCVA w/ SL |
+0.32±0.18 | +0.34±0.18 |
+0.24±0.15 |
|
IC: Irregular Cornea; RC: Regular Cornea; ♀ female; ♂ male; PrimIC: primary ectasia; SecIC: secondary ectasia; LA.RC: Low Astigmatism; HA.RC: High Astigmatism; HC: Habitual Correction; BSC: Best Spectacle Correction; HCVA: High Contrast Visual Acuity; LCVA: Low Contrast Visual Acuity; SL: Scleral Lenses; N/P: No prescription; Q: corneal asphericity; (+) Independent T-test; (
Regarding the results of VA, in IC Group there were statistically significant differences between both HCVA and LCVA with SL when compared to HC (improvement of more than 2 lines, p<0.001). In RC Group those differences were also statistically significant, although clinically insignificant (differences of 2.5 letters, p<0.05). Although HCVA with HC was significantly different between groups, there were no differences between them in HCVA measured with SL, meaning that we can achieve an identical HCVA in healthy and irregular corneas with these kinds of devices. However, there was a statistically significant difference of 1 line of letters in LCVA with SL between the same groups, which reflects that the optical quality in low contrast is significantly worse in subjects with irregular corneas.
Figure
Number of trial lenses required to achieve the best fit. Data is presented in a chronological scale of 20 fittings. Bars represent the mean number of trial lenses and respective standard deviation. Boxes show the median (MED) and interquartile range (IQR) for each chronological group.
The mean number of lenses trialed to arrive to the final dispensing SL in the trial visit was 1.85±0.71 lenses, being 1.84±0.69 for IC Group (range between 1 and 4 trial lenses) and 1.88±0.77 for RC Group (range between 1 and 4 trial lenses). When both groups were compared, there were no statistically significant differences between them regarding the number of trial lenses needed to achieve the best fit (p=0.970, Mann–Whitney U test). By further dividing the sample into subgroups, more lenses, on average, were required for Sec.IC (irregular corneas submitted to surgeries, 1.98±0.72 lenses) than for Prim.IC (1.78±0.67 lenses), but without statistically significant differences between them (p=0.149, Mann–Whitney U test), and more trial lenses for HA.RC (with astigmatism >2.00 D, 1.96±0.82 lenses) than for LA.RC (1.63±0.52 lenses), also without statistically significant differences (p=0.413, Mann–Whitney U test).
Figure
Number of reorders required after the first lens dispensed. Data is presented in a chronological scale of 20 fittings. Bars represent the mean number of trial lenses and respective standard deviation. Boxes show the median (MED) and interquartile range (IQR) for each chronological group.
The average number of reorders needed was 0.76±0.77 lenses, being 0.73±0.76 for the IC Group (range between 0 and 4 trial lenses) and 0.88±0.81 for the RC Group (range between 0 and 3 trial lenses), without statistically significant differences between them (p=0.303, Mann–Whitney U test). By further dividing the sample into subgroups, the Sec.IC required statistically more reorders to achieve the best fit (0.98±0.92) when compared to Prim.IC (0.60±0.63) (p<0.05, Mann–Whitney U test). But when comparing the mean number of reorders between LA.RC and HA.RC (1.00±0.76 and 0.85±0.83, respectively), there were no statistically significant differences (p=0.537, Mann–Whitney U test). However, 73.3% of the reorders performed on RC Group were done on HA.RC subgroup, with also as higher number of fittings (Table
Number of lenses reordered in each group (irregular and regular cornea) and subgroup (surgical/nonsurgical and low and high astigmatism).
TOTAL |
IC Group | RC Group | ||||||
---|---|---|---|---|---|---|---|---|
|
Prim.IC (n=80) | Sec.IC |
|
LA.RC |
HÁ.RC | |||
Cause of reorder |
|
|
|
12 | 15 |
|
3 | 6 |
|
|
(13.5%) | (16.9%) |
|
(10.0%) | (20.0%) | ||
|
|
|
2 | 3 |
|
0 | 2 | |
|
|
(2.2%) | (3.4%) |
|
(0%) | (6.7%) | ||
|
|
|
16 | 5 |
|
0 | 2 | |
|
|
(18.0%) | (5.6%) |
|
(0%) | (6.7%) | ||
|
|
|
5 | 4 |
|
3 | 0 | |
|
|
(5.6%) | (4.5%) |
|
(10.0%) | (0%) | ||
|
|
|
6 | 5 |
|
2 | 8 | |
|
|
(6.7%) | (5.6%) |
|
(6.7%) | (26.7%) | ||
|
|
|
4 | 7 |
|
0 | 3 | |
|
|
(4.5%) | (7.9%) |
|
(0%) | (10.0%) | ||
|
|
|
3 | 2 |
|
0 | 1 | |
|
|
(3.4%) | (2.2%) |
|
(0%) | (3.3%) | ||
|
||||||||
Visit of reorder |
|
|
|
34 | 30 |
|
5 | 16 |
|
|
(38.2%) | (33.7%) |
|
(16.7%) | (53.3%) | ||
|
|
|
3 | 3 |
|
0 | 2 | |
|
|
(3.4%) | (3.4%) |
|
(0%) | (6.7%) | ||
|
|
|
6 | 6 |
|
3 | 4 | |
|
|
(6.7%) | (6.7%) |
|
(10.0%) | (13.3%) | ||
|
|
|
3 | 1 |
|
0 | 0 | |
|
|
(3.4%) | (1.1%) |
|
(0%) | (0%) | ||
|
|
|
2 | 1 |
|
0 | 0 | |
|
|
(2.2%) | (1.1%) |
|
(0%) | (0%) | ||
|
||||||||
|
|
|
48 | 41 |
|
8 | 22 | |
|
|
(53.9%) | (46.1%) |
|
(26.7%) | (73.3%) |
n is the number of fittings;
Most of the reorders performed were due to inadequate sagittal height (more than 30% in both groups), poor vision (23.6%, IC Group), and a combination between poor vision and inadequate fit (33.3%, RC Group). An important issue is that about 10% of the subjects of each group required a reorder because of lens discomfort, although the fitting seemed satisfactory. Most of the changes were done in the landing zone of the lens, namely, refitted with toric designs, which resulted in improved comfort. Another important factor is the number of lenses that broke (5 in IC Group and 1 in RC Group); 4 of them broke during mechanical handling disinfection (rubbing the lenses), 1 lens fell on the floor during application, and 1 lens suffered an
Figure
Percentage of lenses with back toric lens design (a) and front toric lens design (b) required. Data is presented in a chronological scale of 20 fittings.
Several studies have already proven the visual efficacy of SL for different eye conditions, from normal/regular shaped corneas to the more challenging corneal irregularities [
The recently published results of SCOPE online survey on demographic and prescribing patterns of SL fitters [
In the present study, we identified how many trial lenses were required to achieve the optimal lens to be dispensed and how many reorders were required after the first dispensed lenses, and how this learning curve evolves over time in a novel practitioner without previous experience in SL fitting. Figure
Although there are no studies in peer-review literature reporting the potential improvement of practitioner skills over time in fitting SL (learning curve), there are few studies reporting the mean number of trial lenses or lenses ordered per eye to achieve the best fit during the fitting process. Schornack et al. [
The differences between IC and RC groups on the mean number of trial lenses required to achieve the best fit (1.84±0.69 and 1.88±0.77, respectively) were not statistically significant. By further dividing our results in subgroups, postsurgical corneas (Sec.IC) required more trial lenses than those with primary ectasia. Corneas with high astigmatism (HA.RC) also required more trial lenses to achieve the best fit. Although there were no statistically significant differences between them, this means that irregular corneas submitted to surgeries or corneas with high astigmatism could be more challenging to fit in some cases. By personal experience of the practitioner, those corneas that underwent specific surgeries (like penetrating keratoplasty and intracorneal ring implantation) or those with high astigmatic corneas (namely, with limbus-to-limbus high toricity) are often more challenging to fit. Possible explanations to this include the more asymmetric corneal surface in the postsurgical corneas and the more asymmetric scleral shape associated with the highly toric corneas (in the RC Group). Although there is lack of consensus in this regard, some clinical observations revealed that, when the corneal astigmatism is higher and congenital in nature, the sclera could also have the same magnitude and orientation of toricity [
Regarding the reorders needed during the fitting process
Although it seemed to have an increase in the number of reorders in the first fittings, we rapidly see a tendency to decrease (Figure
The differences between both groups on the mean number or reorders (0.73±0. 76 and 0.88±0.81, respectively) were small and with no statistically significant difference. When further dividing into subgroups, and similar to what we concluded about trial lenses, Sec.IC subgroup needed more reorders than Prim.IC (p<0.05), which corroborates our thoughts about the complexity of fitting those corneas that underwent some surgeries. Controversial to the findings on the mean number of trial lenses required in each subgroups of RC Group, no statistically significant differences were found (mean of 1.00±0.76 on LA.RC and 0.85±0.83 in HA.RC). In fact, there is a large difference in the number of subjects of each subgroup (8 in LA.RC and 26 in HA.RC), but we can also see that 73.4% of the total number of reorder were from the subgroup of corneas with high astigmatism. As said before, the clinical feeling of the practitioner was that high astigmatic corneas were more complex to fit. In 3 fittings of HA.RC Group it was required to order a different trial lens because none of the lenses from the trial set fitted correctly the scleral shape (landing zone) because of high scleral toricity.
Regarding the prescribing pattern of more specific designs, 83% of the total fits have toric landing zone designs (85% in IC Group and 74% in RC Group). This is in accordance with Gregory DeNayer et al. [
There are some factors that could be seen as limitations of the study. First, only 1 practitioner/fitter was evaluated to assess the learning curve: other practitioners could learn faster or slower and this will have a direct impact on the study findings. Second, the results of this study are limited to the fitting of SLs using trial sets with the same characteristics of the ones used in this study. Current fitting approaches by most practitioners use a similar procedure what allows to apply current results to most fitting protocols. However, other designs and manufacturers might not replicate exactly the present results and they need to be independently assessed. Also, technologies such as OCT and scleral topographers are increasingly being used during SL fittings, which could aid during the fitting process and consequently decrease the number of trial lens and reorders. Also, techniques derived from corneal topography, like the ones described in another study by the same authors [
The authors’ decision to use both eyes of each subject (when applicable) was because 78% of the total sample were irregular corneas and it is well established that the majority of these conditions are asymmetric in nature. These asymmetries will influence the lens fitting, namely, the lens sagittal height for each eye. In addition, SLs land on conjunctiva, so the anatomy of the eye beyond the corneal borders has an important role in the fitting process. Despite some degree of correlation in refractive error or corneal power between both eyes (which are not necessary related to the difficulty of SL fitting process), there were poor correlations considering the geometry of the lens landing zone in the two groups (r=0.364 IC Group and r=0.333 RC Group, Spearman). Considering the clinical experience of the authors that, despite similarities that might be present between both eyes of the same subject, the level of complexity of the fitting process is not so straightforward; specific adjustments are often required. Further limitations include the asymmetric number of patients/eyes in the different subgroups. However, altogether, the present study presents one of the largest case series recently published.
In summary, we have observed that contemporary scleral supported rigid gas permeable lenses can be fitted in most cases of moderate-to-severe ocular corneal defects and regular shaped corneas by practitioners with minimal previous training. After the first fittings, a novel practitioner would be able to significantly reduce the number of trial lenses and reorders to the manufacturer.
The data used to support the findings of this study are available from the corresponding author upon request.
This project was supported in part by an unrestricted grant from Bausch & Lomb (Rochester, USA) and projects PTDC/SAU-BEB/098391/2008 and PTDC/SAU-BEB/098392/2008 and the Strategic Funding UID/FIS/04650/2013. Authors declare that they do not have any proprietary or financial interest in any of the materials mentioned in this article. Authors thank Procornea (Netherlands) for their valuable cooperation. Part of the present results have been presented in a short communication to the Global Specialty Lens Symposium, GSLS2018, Nevada, USA, 2018. Written consent has been obtained from the patients to publish the information reported in this paper. Datasets are available upon request to the corresponding author.
The authors declare that there are no conflicts of interest.