Article Text
PDF
Abstract
Background/Aims To investigate the efficacy of therapeutic soft contact lenses (SCLs) in gelatinous drop-like corneal dystrophy (GDLD) management.
Methods This was a retrospective, consecutive, observational case series, including 20 patients (40 eyes) with GDLD treated in Osaka University Hospital within the last 15 years. We tested the effects of therapeutic SCL on clinical features, visual acuity and surgical interventions. Examinations for clinical features and visual acuity were done on patients who had no surgical intervention for 3 years. Scoring and evaluation of changes in three main clinical GDLD features and visual acuity (logMAR units) were performed using Fisher’s exact test and Mann-Whitney U test. Surgery-free survival time was compared by Kaplan-Meier analyses in all patients.
Results We found a significantly lower rate of progression in GDLD nodular lesions in patients wearing SCLs compared with those who did not (p=0.0179). No suppressant effects were observed regarding opacity and neovascularisation, and no significant improvements were found in visual acuity (in logMAR values, SCL-on: mean=− 0.036, median=0; SCL-off: mean=0.149, median=+ 0.088; p=0.14). The surgery-free survival time for all 16 SCL-on eyes was 2770 ± 1918 days, significantly longer than that for 22 SCL-off eyes, 1342 ± 1323 days (Kaplan-Meier analysis, p=0.0007), suggesting that therapeutic SCL extends the period until surgical intervention and reduces their necessity in patients with GDLD.
Conclusion Wearing therapeutic SCLs in GDLD slows the progression of nodular lesions and decreases the need for surgical interventions.
- corneal dystrophy
- soft contact lens
- nodular lesions
- visual acuity
- surgical intervention
Statistics from Altmetric.com
Introduction
Gelatinous drop-like corneal dystrophy (GDLD), characterised by the deposition of grey gelatinous masses in the subepithelial space of the cornea, is one of the severest inherited disorders, with an autosomal recessive trait. GDLD is an uncommon disease in Japan, with an estimated incidence of approximately 1:30 000, while few cases have been reported in Western countries.1 2 This is because of a founder effect of the Q118X mutation in the responsible gene, TACSTD2, present in about 90% of the affected chromosomes in Japan.3
With the progression of gelatinous mass deposition, the superficial elevated nodules (called mulberry-type nodules) and the progressive corneal opacity cause blurred vision. Neovascularisation of limbus appears, and the deepened corneal involvement, due to the deposition, gradually causes visual loss.4 The standard treatment is keratoplasty; however, frequent reoccurrences are reported,5 6 and repeated keratoplasty impairs patients’ quality of vision (QOV), while leaving room for possible secondary complications. Therefore, the control of elevated nodules, corneal opacity and neovascularisation, and reduction of surgical interventions are quite important in the management of GDLD.
In patients with GDLD, therapeutic soft contact lenses (SCLs) are often prescribed clinically. Usually, therapeutic SCLs are used for the following purposes: (1) to support and protect the cornea,7 (2) to manage pain of the orbital surface in bullous keratopathy,7 8 (3) to aid in epithelial healing following abrasion or recurrent corneal erosion after refractive surgery or therapeutic intervention involving the ocular surface and hospitalisation for burns or trauma,7 8 (4) as care and treatment for persistent corneal epithelial defects7 8 and (5) for maintenance of the anterior chamber after corneal perforation.8 In GDLD, therapeutic SCLs are not generally prescribed in clinical practice for these reasons, but are based on the clinical impression that it improves GDLD clinical features, as reported in a previous case report.9 However, none of these previous reports have analysed the efficacy of therapeutic SCLs quantitatively and in multiple cases, and no evidence has been provided for the efficacy of wearing SCLs. The purpose of this study was to quantitatively evaluate the efficacy in the management of GDLD, by examining records of all patients with GDLD in Osaka University Hospital over a period of 15 years.
Methods
This was a retrospective study, conducted in adherence to the tenets of the Declaration of Helsinki.
Twenty patients (40 eyes) who presented with GDLD at Osaka University Hospital between February 2002 and September 2017 were enrolled in the study. In these patients, we studied the effect of long-term SCL wearing on clinical outcomes, including clinical features, visual acuity and surgical interventions (see below for detailed definition of each outcome). The baseline information of all patients is shown in online supplementary table 1.
Supplemental material
We applied hydrogel SCLs of continuous use (Plano T (BAUSCH+LOMB Ireland), Breath-O (Toray Industries), Plano B4 (BAUSCH+LOMB Ireland), Acuvue (Johnson&Johnson Vision Care) or Medalist Plus (BAUSCH+LOMB Ireland)) or a silicone–hydrogel contact lens of continuous use (AIR OPTIX EX AQUA (Alcon)) (online supplementary table 2). As sale periods varied for different SCLs, the selected SCLs were not the same in all individuals: Plano T was applied in 3 eyes, Breath-O in 2, Plano B4 in 15, Acuvue in 31, Medalist Plus in 4 and AIR OPTIX EX AQUA in 20 eyes.
Supplemental material
We performed three separate analyses to examine the effects of SCL wearing in patients with GDLD: analysis of (1) clinical features, (2) visual acuity and (3) the need for surgical interventions.
Clinical features
We studied SCL wearing effects on three prominent GDLD clinical features: nodularity, opacity and neovascularisation.10 To exclude surgical effects, we selected 17 patients who had not undergone any surgical procedure for 3 years from their first visit and who had either continuously worn or never worn SCL in those initial 3 years. Surgical procedures included superficial corneal resection, phototherapeutic keratectomy, anterior lamellar keratoplasty, deep anterior lamellar keratoplasty, penetrating keratoplasty, limbal transplantation and keratoepithelioplasty. One patient who had trauma during the first 3 years was excluded.
As no evaluation method has been reported for the severity of these clinical features, we generated a new grading scale for this study. A score of 0 to 3 was allocated to each of the following three criteria: nodular lesions, corneal opacity and neovascularisation (figure 1). For each feature, scoring was performed based on the number of grey raised masses, the area of cornea covered with opacity with a modified form used for corneal haze grading11 12 or the amount of vascular invasion in the limbus observed.
Scoring system to evaluate three clinical features of gelatinous drop-like corneal dystrophy: nodular lesions, corneal opacity and neovascularisation. Four-point (0 to 3) scoring systems were used for grading. The grading was performed independently by two blinded corneal specialists based on ocular photographs of the anterior segment, in a retrospective manner. In case the grading was different between the two specialists, a third specialist determined the grade in this study. For each feature, scoring was performed based on the number of raised masses, the area of cornea covered with opacity or the amount of neovascularisation observed. The scores for nodular lesions were based on the number of raised grey gelatinous masses, with 0 used for ‘no depositions’; 1 for ‘1 to 5 nodules’, 2 for ‘6 to 10 nodules’ and 3 for ‘more than 10 nodules’. The score for corneal opacity was based on the area of opacity, with 0 used for ‘clear cornea’, 1 when ‘all pupillary area is visible’, 2 when ‘partial pupillary area is visible’ and 3 when ‘pupillary area is invisible’. The score for neovascularisation was based on vascular invasion, with 0 used for ‘none’, 1 for ‘one quadrant of neovascularisation’, 2 for ‘two quadrants of neovascularisation’ and 3 for ‘more than three quadrants of neovascularisation’ in the limbus.
The eyes were assigned to two study groups based on continuous SCL wearing (SCL-on group) or no SCL wearing (SCL-off group) over 3 years. SCL effects on the above three features were evaluated by comparing the baseline score to that after 3 years. Decreased scores indicated ‘improvement’, no change in scores indicated a ‘progress restraint’ and increased scores indicated ‘aggravation’ (online supplementary figure 1). Improvement and progress restraint were defined as ‘lower rate of progression’. There were cases in which the grading score was 3 both at baseline and at the 3-year visit. As we did not consider adequate to evaluate the severity of cases with such severe pre and post states, we classified them as ‘not applicable’ and excluded them from the analysis.
Supplemental material
Visual acuity
Changes in visual acuity between the SCL-on and SCL-off groups were compared. As in the case of clinical features, to exclude surgical effects, we selected patients who had not undergone any surgical procedure in the last 3 years from their first visit and those who met the condition of continuously wearing (SCL-on group) or not wearing SCLs (SCL-off group) in those 3 years. Thus, 17 eyes remained for this comparison and were assigned to two study groups, as mentioned above. The visual acuities before and after 3 years were evaluated in LogMAR units.
Surgical interventions
The time from the first visit to the endpoint was compared between patients in the SCL-on and SCL-off groups. Among the 40 eyes, two were excluded from the analysis due to lack of essential clinical information, one due to lack of monitoring points because the patient was followed-up by another clinic and another one due to unknown surgery date because the surgery was performed in another clinic. The starting point of monitoring was set as the date of the first visit to our hospital where the visual acuity was recorded. The endpoint was the date on which the patients underwent a surgical procedure, as defined above. Failure of clinical visits and transference to other clinics were considered as censored observations.
Statistics
Statistical comparisons of demographic characteristics were made between the SCL-on and SCL-off groups using Student’s t-test and Fisher’s exact test for numerical and non-numerical data, respectively. Fisher’s exact test was performed to determine the differences in the clinical score between the two groups. Changes in visual acuity over 3 years were compared between the two groups by two-sample Wilcoxon rank-sum (Mann-Whitney) test. The surgery-free survival time was compared between groups by Kaplan-Meier analysis, assessed by log-rank and Wilcoxon tests. Data were considered statistically significant at p<0.05.
Results
SCL effects on clinical features
Among all patients, 17 eyes of 11 patients were included. Among them, 11 eyes had continuously worn SCL (SCL-on group: 7 males, 4 females), and 6 eyes had not (SCL-off group: 3 males, 3 females). The patients’ clinical features at enrolment (first visit) are shown in online supplementary table 3. In patients of the SCL-off group, the median scores for nodular lesions, corneal opacity and neovascularisation at the starting point were 0.5, 2.0 and 2.5, respectively. In patients of the SCL-on group, the scores were 0, 1.0 and 1.0, respectively. There was no significant difference at the starting-point scores for nodular lesions (p=0.534) or neovascularisation (p=0.251) between the groups; however, the baseline score for corneal opacity was higher in patients of the SCL-off group (p=0.004).
Supplemental material
Clinical improvement of nodular lesions was apparent (online supplementary figure 2) in patients of the SCL-on group, compared with those of the SCL-off group (p=0.0179). A representative case is shown in figure 2. After 3 years, a lower rate of progression (maintained or improved) was observed in all cases (11 of 11, 100%), and there was no aggravation in any patients of the SCL-on group. In contrast, only 33% of the SCL-off cases showed a lower rate of progression regarding nodular lesions.
Supplemental material
A representative case showing the effect of therapeutic soft contact lens (SCL) usage for 3 years. Image of the right eye of a 43-year-old female with gelatinous drop-like corneal dystrophy before using SCL (A). Note the prominent nodules of score 3 (more than 10 nodule lesions, arrowheads). Scores for nodules, corneal opacity and neovascularisation were 3, 1 and 1, respectively. image after wearing therapeutic SCL for 3 years (B). An apparent improvement in nodular lesions is observed. Scores for nodules, opacity and neovascularisation were 0, 1 and 1, respectively.
For further confirmation, we also tried to assess nodular lesions by anterior segment optical coherence tomography (AS-OCT). Since most patients were enrolled before AS-OCT was well developed, it was difficult to point out each nodular lesion in AS-OCT. However, with SCL, we observed decreased irregularity and smoothening of the corneal epithelium, possibly indicating decreased nodular lesions of grey gelatinous masses in the subepithelial space of the cornea (online supplementary figure 3). For opacity and neovascularisation, there were no suppressant effects of SCL wearing (p=0.509 and p=1.000, respectively; table 1). Moreover, based on the clinical records, the subjective symptoms, such as photophobia, foreign body sensation and epiphora, were reduced by SCL wearing in 3 of 11 eyes in the SCL-on group; for the other 8 eyes, there were no subjective data in the clinical records.
Supplemental material
Results and effects of therapeutic SCL in three clinical features
When we confined the analysis to three patients (six eyes) who wore SCL in one eye but not in the other for 3 years, all three SCL-off eyes received a nodule aggravation score of +1, compared with the three SCL-on eyes that showed no change in the nodule score. However, two SCL-on eyes received a neovascularisation aggravation score of +1, compared with SCL-off eyes, which showed no change in the neovascularisation score. In fact, SCL wearing had a lower rate of progression on nodular GDLD lesions, but not on corneal opacity or neovascularisation.
SCL effects on visual acuity
To address whether SCL can interfere with visual impairments in GDLD management, we compared changes in visual acuity in patients using or not using SCLs. Similar to clinical features, surgical interventions are known to strongly affect visual acuity. Thus, we analysed the same patient subgroups as described above, including patients who had not undergone surgical intervention for 3 years.
The mean and median change in visual acuity (logMAR) was −0.036 (SD: 0.299) and 0 (IQR: −0.301 to 0), respectively, in the SCL-on group, and 0.149 (SD: 0.209) and 0.088 (IQR: 0 to 0.301), respectively, in the SCL-off group. These results indicated that SCL usage slightly improved or maintained visual acuity during the 3 years; in contrast, this effect was not observed when SCLs were not used. However, there was no statistically significant difference between groups (p=0.14).
Next, we confined the analysis to the three patients (six eyes) who used SCL in one eye but not in the other for 3 years. The visual acuity (logMAR) in the three SCL-off eyes worsened or did not change; the mean change in each eye was 0.477, 0.000 and 0.301. However, the visual acuity in the three SCL-on eyes either improved or did not change; the mean change in each eye was 0.000, –0.301 and 0.000. Despite this effect, there were no significant differences between groups (p=0.10). Thus, we could not confirm whether SCLs have a suppressive effect on visual acuity impairment.
SCL effects on surgical interventions
The mechanism by which nodular lesion progression affects quality of life (QOL) or QOV is not understood; thus, we decided to assess the direct benefits of SCL in our patients. As GDLD affects the corneal epithelium,13 it frequently recurs after keratoplasty.5 To address this, we examined whether SCL usage reduced the need for surgical intervention. This could improve patients’ QOL.
In this analysis, we considered SCL usage effects on surgical interventions in all GDLD eyes in our clinic, presenting within 15 years and considered surgical intervention as the endpoint. Accordingly, 16 eyes were included in the SCL-on and 22 eyes in the SCL-off group. Among the 38 eyes, 27 underwent some surgical intervention during the 15 years of observation (online supplementary table 4). Next, we compared the surgery-free survival time, defined as the period of monitoring, in 11 cases that did not have any surgical intervention (9 SCL-on and 2 SCL-off eyes) or as the period from the first clinic visit to the time of surgical intervention in 27 cases that had one or more operations within the monitoring period (9 SCL-on and 18 SCL-off eyes). There was no significant difference in the score of nodular lesions or corneal opacity (Wilcoxon rank-sum test, p=0.060 and 0.065, respectively). The baseline score for neovascularisation was higher in the SCL-off group (Wilcoxon rank-sum test, p=0.031).
The rate of surgical intervention for SCL-on and SCL-off eyes were 4.57 and 25.03 per 100 person-years, respectively. Surgery-free survival time for all 38 eyes was 1944±1731 days (median: 1486 days). The surgery-free survival time for the 16 SCL-on eyes was 2770±1918 days, which was more than double that of the 22 SCL-off eyes (1342±1323 days; table 2). Significant differences between the SCL-on and SCL-off groups were determined based on Kaplan-Meier curves (figure 3). We found that the overall surgery-free survival time was significantly longer in the SCL-on than in the SCL-off group, as assessed by log-rank and Wilcoxon tests (p=0.0007 and p=0.0066, respectively), suggesting that the use of therapeutic SCLs extends surgery-free survival and reduces the need for surgical intervention in patients with GDLD.
Average and median surgery-free survival time for SCL-on and SCL-off in the third analysis
Kaplan-Meier curves plotting the surgery-free survival time for the soft contact lens (SCL)-on and SCL-off groups. The starting time of the follow-up was set as the date of the first visit to our hospital, and the time-to-event was defined as the time between the starting time and the date on which the patients underwent a surgical procedure. We defined “failure” as having a surgical procedure, therefore, ‘survival’ was considered as an avoidance of surgical procedures. Significant differences are seen between groups, as assessed by log-rank and Wilcoxon tests (p=0.0033 and p=0.0186, respectively).
Discussion
In our study, we investigated the effectiveness of SCL usage in patients with GDLD. A previous case report showed an improvement in the clinical appearance and subjective symptoms of GDLD9 with a qualitative description; however, no evidence has been yet provided on the appropriateness of SCL usage in GDLD management. To address this, we examined all patients with GDLD who had visited our hospital over a period of 15 years from three perspectives: (1) clinical features, (2) visual acuity and (3) the need for surgical interventions.
In our study, SCL usage clearly improved nodular lesions, with no significant effect in corneal opacity and neovascularisation. The different outcomes among these three features may have resulted from differences in the respective scoring systems used for their assessment: scoring of nodular lesions was done quantitatively, by measuring the numbers of deposits, which is completely objective; in contrast, scoring of corneal opacity and neovascularisation was done qualitatively by estimating the extent of areas affected. Corneal opacity and neovascularisation may only slightly progress or be affected by SCL usage. Thus, it was difficult to observe significant changes in the respective scores in the 3-year observation period. SCL usage clearly improved nodular lesions in GDLD, as well as subjective symptoms, such as photophobia, foreign body sensation and epiphora.
The mechanism by which SCLs might improve nodular lesions is unknown. It has been demonstrated that deleterious TACSTD2 mutations lead to loosening of cell-to-cell tight junctions due to the lack of claudin-1, ZO1 and occludin in surface epithelial cells.14 15 In GDLD, this deficiency in epithelial barrier function allows the invasion of excessive protein that permeates into the corneal stroma due to increased tear permeability.13 However, the detailed mechanism of the accumulated protein degenerating into amyloid nodular deposits is still unknown. Presumably, SCL usage might influence the invasion of excessive proteins. For example, continuous therapeutic SCL usage might (1) decrease the turnover and fall-off of surface epithelial cells; (2) decrease the turnover of tear fluid around the corneal surface16; (3) protect from mechanical stimulation stress, such as blinking9 16 or (4) capture lactoferrin, factors that are all considered to decrease the permeation of the tear fluid into the corneal tissue and contribute to reduced subepithelial precipitation in GDLD.
GDLD is an autosomal recessive Mendelian disease, caused by the loss of function of TACSTD2, which results in dystrophy.3 Among the 40 eyes included in our study, mutational analyses were performed in 28: the mutational status of the other 12 eyes remains unknown. However, we detected the p.Gln118Ter mutation in 22 eyes (79%), the p.Gln211ArgfsTer60 mutation in 4 eyes (14%) and the p.Ile156Asn in 2 eyes (7%). Consanguineous marriage was found in 18 eyes (45%), none in 8 eyes and was unknown in 14 eyes. The p.Gln118Ter mutation was the most common alteration in our patients, consistent with other reports including Japanese patients.1 The phenotypes of the 40 eyes included in our study at first visit were classified into four distinct subtypes: band keratopathy (9 eyes), stromal opacity (16 eyes), typical mulberry (15 eyes) and kumquat-like type (0 eyes); nevertheless, a previous operative history could have influenced these phenotypes of recurrent GDLD. Eyes with no prior surgery were 14 in total, 7 eyes with band keratopathy, 1 with stromal opacity and 6 with typical mulberry. In previous literatures, genotype–phenotype correlations have not been observed in GDLD, and phenotypes are variable despite the allelic homogeneity of p.Gln118Ter.17 18 Although we could not determine the genotypes of all enrolled patients, there was no significant difference in the rate of p.Gln118Ter mutation between the SCL-on and SCL-off groups; thus, our results should not be biased by the genotype. In countries outside Japan, mutations other than the p.Gln118Ter, such as the p.Glu227Lys, were reported,10 and the potential utility of therapeutic SCL may equally well be evident for GDLD management outside of our Japanese genetic cohort.
The efficacy of SCLs was apparent regarding surgical intervention, which is quite important for the patients’ QOL. Repeated keratoplasty leaves room for possible secondary complications, such as glaucoma, cataract, infections and rejection of the corneal graft.10 Steroid-induced glaucoma frequently and severely occurs in GDLD, and it is speculated that amyloid depositions may also exist in the trabecular meshwork of patients with GDLD, with an inclination towards increased intraocular pressure.16 As our study was retrospective, the quality of evidence may be limited. For example, in severe cases, clinicians may decide to operate immediately and, thus, may not consider prescribing SCLs. To exclude such bias, we performed an additional analysis based on Kaplan-Meier curves, by excluding cases for which surgeries were decided and scheduled on the first clinical visit. Despite the aforementioned disadvantageous bias, the overall surgery-free survival time for SCL-on eyes was still significantly longer than that for SCL-off eyes, assessed by the log-rank test (p=0.0052; online supplementary table 5 and figure 4). The reason for this effect is not entirely clear. However, based on our results and those of previous reports,9 it seems that SCL wearing reduces strong subjective symptoms and may reduce surgical interventions.
Supplemental material
Supplemental material
Based on these results, we conclude that SCL usage has positive effects in GDLD management, as it has a lower rate of progression on nodular lesions and reduces the need for surgical interventions. Our results indicate that SCLs may constitute a powerful and recommendable therapy in the management of GDLD.
Supplemental material
References
Footnotes
Contributors SM, TS, MT and RS designed and directed the project and interpreted data. SM and MT wrote the manuscript. TS, RK, YO, SK, KM, SK, NM and KN revised the manuscript and provided critical advice regarding discussion. Additionally, all authors discussed the results and commented on the manuscript.
Funding This work was supported by JSPS KAKENHI Grant Number JP 17K11448 to Tsujikawa.
Competing interests During the conduct of the study, MT received grants from Japan Society of Promotion of Science and NM received personal fees from Alcon. MT reports personal fees from Santen and Alcon, outside the submitted work. TS reports personal fees from HOYA, Santen, Otsuka, Nitto Medic, Pfizer and Senju, outside the submitted work. RK reports grants and personal fees from Novartis, Pfizer, Senju and Bayer; personal fees from Alcon, Kowa and Santen and other fees from Topcon, Predictive analytics, Office Future, and MICIN, outside the submitted work. YO reports personal fees from Senju, Santen, Alcon Japan and Otsuka, outside the submitted work. KN reports personal fees from Alcon, Menicon, Oculus, Otsuka, Santen and Seed; grants and personal fees from Johnson & Johnson; and grants from Shire, outside the submitted work. NM reports grants from Topcon, the Japanese Ministry of the Education, Culture, Sports, Science and Technology, and the Japanese Ministry of Health, Labor and Welfare; personal fees from AMO, Alcon, B+L, HOYA, Oculus, Santen, Senjyu, Tomey, Menicon and Pfizer, outside the submitted work. KN reports personal fees from Senju, Otsuka, Japan Tissue Engineering, Santen, Alcon, Carl Zeiss, HOYA, Johnson & Johnson, Novartis, Pfizer, Wakamoto, Seed and Nidek, outside the submitted work.
Patient consent for publication Not required.
Ethics approval The study was approved in advance by the Institutional Review Board/Ethics Committee of Osaka University.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- At a glance