Scleral lens wear: Measuring inflammation in the fluid reservoir

Abstract

Purpose:

To measure inflammatory mediators in the scleral lens fluid reservoir (FR) in healthy eyes and to compare them to basal tear samples after 8-hs (8h) and 4-days (4d) of scleral lens (SL) wear.

Methods:

Fifteen normal, habitual soft contact lens wearers were fitted with 14.8- or 15.4-mm SLs (Zenlens, Alden Optical, USA). Basal ocular surface tears and FR samples were collected after 8h and 4d of daily SL wear. Levels of interleukin (IL) -4 and -8, matrix metalloproteinase (MMP)-7, -9, and -10, and tissue inhibitor of MMPs (TIMPs) 1–4 were measured in all samples using Luminex assays. Visual acuity, corneal and conjunctival staining, and comfort assessments were completed at the baseline, 8h and 4d time points.

Results:

MMP-9 and MMP-10 were greater in FR than basal ocular surface tears. After 8h of SL wear, the median concentration of MMP-9 in the FR and basal tears were 62.7 and 15.2 ng/mL, respectively (p = 0.047). Likewise, MMP-10 was significantly greater in FR compared to basal tears, after 8h (25.8 ng/mL vs 2.8 ng/mL, p < 0.001) and 4d (2.1 ng/mL vs17.2 ng/mL, p = 0.047). IL-4 and IL-8 levels were greater in FR but not significantly at 8h (2.2 vs 3.1 ng/mL; and 0.1 vs 0.4 ng/mL, respectively) or 4d (0.9 vs 3.5 ng/mL; 0.0 vs 0.2 ng/mL). MMP-7 was not affected by SL wear after 8h (46.0 basal vs 54.4 ng/mL FR) or 4d (34.2 vs 87.5 ng/mL). Visual acuity, corneal and conjunctival staining did not change; comfort was reduced in SL compared to soft contact lens wear.

Conclusions:

This is the first study to compare the FR with the basal ocular surface tears. MMP-9 and MMP-10 were elevated in the FR after several hours of SL wear, suggesting potential clinical implications of SL wear and deserves further investigation.

1. Introduction

The scleral lens (SL) is an ocular surface device manufactured in gas-permeable plastic and placed on the eyes of individuals with corneal disease. Originally manufactured in glass and used to treat high myopia and irregular corneas in Europe during the late 19th century, it wasn’t until the development of gas-permeable materials in the late 20th century that the SL became a widely viable option; in the past 15–20 years the gas-permeable SL has dramatically integrated into clinical practice. Modern applications of the SL remain to manage irregular corneal disorders like keratoconus [1–3] and have expanded to include s/p corneal surgeries [4,5] and the treatment of ocular surface compromise caused by Sjögren’s syndrome [6,7], ocular cicatricial pemphigoid [8,9], graft-versus-host disease [5,10], and other environmental exposure-related or dry eye diseases [1,7,11]. Over the past two decades, the use of the SL has soared with the global realization of the superior comfort and visual stability it can provide [7,12,13].

The expanding use of the SL has led to an increase in case reports that establish the benefits of their use [14–17], but also in the reports of clinical conundrums that raise important questions about the side effects of the SL on the ocular surface [17–26]. The need for additional prospective research focused on the impact of SL on ocular health is recognized by both researchers and clinical practitioners [18,27,28], and the basis of the need is the unique SL fit, which can lend way to several ocular surface sequelae that are not observed with any other ocular medical devices. When a SL is fit to the ocular surface, the large diameter, relatively thick (~300 μm) plastic lens vaults over the cornea, landing on the conjunctiva and harboring a relatively thick post-lens tear fluid reservoir (FR) between the SL and the cornea. Prior to application, the concave portion of a SL is filled with a preservative-free solution (customarily saline), which mixes with the ocular surface tears during SL application to form the FR. This layer is often considered beneficial to the cornea as a protective fluid barrier from the environment, also helping to neutralize higher order aberrations seen in patients with irregular corneal shape. The FR is commonly between 200 and 400 μm in axial depth (thickness), with an estimated volume of approximately 200–400 μl. During SL wear, the FR often has minimal exchange with the outer basal surface tears [22,29–32], meaning that the SL may sequester inflammatory and other tear film components in the FR that would otherwise be refreshed during blinking. Several complications associated with SL wear do in fact occur within the FR, such as midday fogging [14,33,34] epithelial toxicity [4], and underlying corneal edema [35–37]. A recent study reported the presence of leukocytes in the FR of SL wearers experiencing midday fogging suggesting inflammation is occurring on the ocular surface [34]. The composition of the FR and how it differs from the basal ocular surface tear film is otherwise unknown.

The current evidence supports the hypothesis that SL wear traps normally refreshed tear film components in the FR leading to elevated levels of inflammatory mediators overlying the cornea and perilimbal conjunctiva. To test this hypothesis, we determined if pro-inflammatory cytokines (Interleukins: IL) and matrix metalloproteinases (MMPs), which are often used as biomarkers for inflammation in the tear film [38–42], are increased in the FR after 8hours and 4 days of SL wear in normal individuals.

2. Methods

2.1. Participants

This study was compliant with the tenets of the Declaration of Helsinki and was approved by the University of Houston’s Institutional Review Board. All enrolled subjects signed an informed consent prior to participation. A total of sixteen normal, habitual soft contact lens wearers (SL neophytes) were recruited and seen at the University of Houston, College of Optometry (UHCO). Inclusion criteria was daily soft contact lens wear to ensure that all eyes were accustomed to a contact lens being applied to the eye (spherical, soft, or multifocal power). Subjects with a history of extended wear soft contact lens wear were excluded, as with those with a history of corneal gas permeable or hybrid contact lens wear.

2.2. Scleral Lens fit and wearing schedule

Subjects reported to The Ocular Surface Institute (TOSI) at UHCO for a complete anterior segment examination which included assessment of the cornea, conjunctiva, sclera, eyelids and lashes, anterior chamber, and iris. All subjects were determined to have good ocular health and were then fitted with SLs. A diagnostic fitting set with 14.8- and 15.4-mm SLs was used to determine the lens parameters for each subject (Zenlens RC, Alden Optical, Rochester, NY, USA) and selection of diameter was based on horizontal visible iris diameter (HVID). Subjects were fitted into 14.8 mm lenses if their HVID was <11.8 mm, and 15.4 mm lenses when their HVID was ≥11.8 mm. Sagittal depth (SAG) was determined by applying various diagnostic lenses and determining which lens vaulted the apical cornea nearest to 300 μm. During the diagnostic fitting, the investigator also examined the transition zone radius (TZR) and the landing zone radius (LZR), two peripheral lens areas overlying the limbus and the sclera, respectively, to determine if the lens was adequately vaulting the cornea and landing evenly on the conjunctiva. Toric TZR and LZR designs were ordered when meridional asymmetries of compression or excess lift were observed. An ideal SL fit vaulted over the cornea, clearing the limbus by approximately 30–50 μm and landing on the conjunctiva without impingement of blood vessels. Over-refraction was conducted to determine best SL power. Custom SL were ordered and finalized for each subject prior to beginning the experimental visits.

Subjects were instructed to discontinue soft contact lens wear for a 3-day washout period prior to the start of the SL experimental visits, to allow the eye to return to a relative baseline state and avoid interference of any inflammation caused by a soft contact lens fit. On the morning of the first experimental day, baseline testing was done (ocular surface tear collection, comfort, vision and ocular health), and the SL were dispensed between 7:30 and 8:30 AM to be worn continuously before returning for the 8-hour (8h) follow-up visit that evening. At the 8h visit, all initial testing was repeated, and in addition the FR was collected during SL removal. At the completion of the 8h visit, subjects wore the SL at least 8h per day for 3 consecutive days, returning after 8h on the 4th day for the 4-day (4d) visit to evaluate a potential adaptive response. During this wearing period subjects were instructed to remove the SL at night and use a hydrogen peroxide disinfection and cleaning solution to disinfect SL daily (ClearCare^®, Alcon Laboratories, Ft Worth, TX, USA). Each morning the SL was filled with sterile saline solution prior to application (Purilens, Freehold, NJ, USA).

2.3. Tear collection

Basal ocular surface tears (T_b) were collected using a 10 μl microcapillary tube placed in the lower temporal fornix, taking care to avoid the eyelid margin and reflex tearing. Capillary action facilitates movement of tears into the tube. The T_b were collected at the baseline visit (prior to SL wear) and at the 8h and 4d visits (T_b samples at 8h and 4d were collected prior to SL removal). The FR samples (T_FR) were collected as the SL was removed by the investigator (using a micropipette) (Fig. 1). For each sample type (basal, FR), left and right eyes were pooled for each subject (to maximize FR volume) in a single Eppendorf tube and frozen at −80 C until analysis. In total, five tear samples (three T_b and two T_FR) were collected for each subject. No samples were pooled between subjects.

Fig. 1.

Basal tear and fluid reservoir collection. A microcapillary tube was placed in the temporal fornix to collect basal tears (T_b) from the exposed ocular surface prior to SL removal when applicable (A). The SL was carefully removed using a small plunger (B) and the tear fluid reservoir (T_FR) was collected from the lens basin using a pipette (C).

2.4. Cytokine and MMP luminex assay

All tear samples (2–140 μl) were frozen immediately after collection and thawed at the time of analysis. Two- microliters of undiluted sample were used to determine total protein concentration by the use of the Direct Detect^® infrared spectrometer (EMD Millipore, San Diego, CA, USA). Levels of IL-4, IL-8, MMP-7, MMP-9, and MMP-10 were quantitated in tear samples using customized magnetic beads-based Luminex assays (R&D Systems, TC. Minneapolis, MN, USA). All assays were performed according to manufacturer instructions. Briefly, standards, quality controls and samples were pipetted into individual wells of a 96-well plate and thoroughly mixed and incubated with antibody-immobilized beads at room temperature for 2 h. Then, a cocktail of biotinylated detection antibodies specific to the analytes of interest was added to all wells, thoroughly mixed and incubated for 1 h. Next, development was done by adding a streptavidin-phycoerythrin (SAPE) conjugate, which was thoroughly mixed into each well and incubated for 30 min at room temperature. Each incubation step was followed by proper washing to remove unbound sample components or reagents. Finally, SAPE-analyte-binding magnetic beads were re-suspended in sheath fluid and the 96-well plate was analyzed with a MAGPIX instrument and xPONENT software (Luminex Corporation, Austin, TX, USA). Quantitation of each analyte per sample was determined using the Milliplex Analyst software (EMD Millipore). For all Luminex assays, a total of 10 μg of total protein was loaded per well in technical duplicates or triplicates. Therefore, to calculate the final analyte concentration in each sample, the individual dilution factor (that resulted from each sample being diluted to reach 10 μg of total protein per well) was applied, and it is shown as ng/mL for all analytes. The volume of undiluted sample used per Luminex assay ranged from 2 to 12 μl depending on its total protein concentration which varied from 1.80 to 12.97 μg/μl.

2.5. Comfort & ocular health evaluation

Comfort, visual acuity, and ocular surface staining were quantified at baseline and follow-up visits. Contact lens discomfort is the primary reason for discontinuation of contact lens wear in the US [43], and was measured as an assessment of SL satisfaction using two surveys: the Contact Lens Dry Eye Questionnaire (CLDEQ-8) and a custom Visual Analog Scale (VAS). The VAS asked participants to rate their ocular comfort on a 100 mm scale (the left-most limit of the line indicating the SL “extremely uncomfortable”, and the right-most limit indicating “extremely comfortable”). Vision was assessed using a high contrast, high luminance logMAR visual acuity chart. Visual acuity was measured with habitual spectacle wear at baseline, and with SL at initial application, 8h, and 4d.

Bulbar conjunctival staining was measured in 4 quadrants (nasal, temporal, inferior, and superior) after instillation of Lissamine Green (Green Glo, HUB Pharmaceuticals, Rancho Cucamonga, CA, USA). A modified NEI staining scale of 0–3 was used for each quadrant, with a potential total score ranging from 0 to 12. Corneal staining was graded after instillation of sodium fluorescein (Soft Glo, HUB Pharmaceuticals) in a total of 5 corneal areas (central, nasal, temporal, inferior, and superior) using a modified Oxford grading system of 0–5 in each area for a total possible score of 25.

2.6. Statistics and data analysis

Statistical analysis was performed using GraphPad Prism 7.0 (GraphPad Software, La Jolla, CA, USA). The D’Agostino-Pearson omnibus K2 normality test was used to determine normality, and tear analytes were compared using one-way ANOVA and Friedman test for multiple comparisons of non-parametric data. The non-parametric uncorrected Dunn’s test was used for post-hoc multiple comparisons.

Sample size was determined based on feasibility of the study and to collect pilot data, since no preliminary data was available on inflammatory analyte levels in the FR of a SL. Therefore, a sample size of 15 was determined with a goal of 12 complete subjects, the recommended sample size when little is known about the expected outcome [44]. However, due to limited volume of several FR samples, only 10 subjects that had complete datasets and were analyzed. A minimum of 10 individuals was considered acceptable as it has been reported by previous similar pilot studies [44–46], and given the novelty of this type of data. Post-hoc sample size analysis was done using the outcomes from MMP-10 and show post-hoc power of 77.2 % for this data.

3. Results

A total of 16 subjects were recruited for this study and 15 completed all study visits (one subject relocated before completing). However, data is only shown for the 10 subjects with a complete collection of tear samples. The mean subject age was 26 years (range 22–29 years) and 60 % (n = 6) were female. On average, subjects wore the SL for 8.1 ± 0.2 h on the first day of wear, approximately 8h per day on the 2nd and 3rd days, and 8.5 ± 0.4 h on the 4th and final day of SL wear. All subjects reported wearing the SLs for at least 8h each day during the 4 day study period.. The study population demographics are shown in Table 1.

Table 1.

Study population demographics, SL parameters, and central clearance values; shown as mean ± SE unless otherwise indicated.

Subject Demographics
Age (range)	26 (22–29)
Gender (% female)	60 %, n = 6
Hours SL wear: Day 1	8.1 ± 0.2
Hours SL wear: Day 4	8.5 ± 0.4
SL Parameters
Brand	Zenlens RC & Toric RC
Manufacturer	Alden Optical, B&L
Material	Boston XO2
Dk, barrer	141
Power range, diopter	+1.00 to −7.75
Diameter, mm	14.8 and 15.4
SAG range, μm	3600 to 4500
Altered TZR, # lenses (%)	1/20 (5%)
Altered LZR, # lenses (%)	7/20 (35 %)
Central SL Clearance
Apical Clearance (at dispense), μm	293 ± 41
Apical Settling (8h), μm	145 ± 30
Apical Settling (4d), μm	140 ± 32

SL: scleral lens; SAG: sagittal depth; TZR: transition zone radius; LZR: landing zone radius.

3.1. SL fitting characteristics

Two subjects (4 eyes) were fitted into 14.8 mm SL and the remaining 8 subjects (16 eyes) were fitted in 15.4 mm SL. One out of 20 SL (5%) was steepened in the TZR to increase clearance over the limbus; no toric TZR were ordered. In the LZR overlying the conjunctiva and sclera, a total of 7 SL (35 %) were designed as toric, and the remaining 14 lenses (70 %) had a spherical LZR. Mean apical clearance over the center of the cornea was 293 ± 41 μm at dispense, settling at 145 ± 30 μm on day 1 after 8h, and 140 ± 32 μm after 4d of SL wear (Table 1). Average apical clearance after 8h of SL wear on day 4 was 157 ± 88 μm.

3.2. Total tear protein analysis

A complete set of 5 tear samples were required to test whether the concentrations in the T_b were different than in the T_FR. Full sample sets were collected in 10 subjects, with those excluded having one or more missing either T_FR (n = 4) or T_b (n = 1) samples. Mean T_b volume collected was 17 ± 2 μL prior to SL wear and 18 ±1 μL after SL wear; mean T_FR volume collected was 30 ± 5 μL. Total protein concentration (TPC) was greatest in the T_FR samples (7.8 ± 0.5 μg/μL) but it was not significantly different from the concentration of the baseline tears collected prior to SL wear (6.2 ± 0.7 μg/μL) or the T_b samples taken prior to SL removal (5.8 ±0.6 μg/μL) (p = 0.14) (Table 2).

Table 2.

Tear collection volumes and total protein concentration (TPC). A total of five tear samples were collected from each subject. Baseline basal tears (T_b) were collected prior to SL fitting and dispense. The basal and tear film reservoir (T_FR) tear samples were then collected after 8h and 4d of SL wear. Data shown as mean ± SE.

	Baseline	8h		4d
	Tb	Tb	TFR	Tb	TFR
Volume (μl)	17 ± 2	19 ± 2	24 ± 4	18 ± 1	36 ± 12
TPC (μg/μl)	6.2 ± 0.7	5.7 ± 0.6	7.0 ± 3.1	5.9 ± 3.5	7.8 ± 2.5

3.3. Cytokine and MMP luminex assay

To determine pro-inflammatory cytokines on the ocular surface during SL wear, we measured and compared the concentrations in the T_b and T_FR samples, at both 8h and 4d. The two cytokines tested in all subjects, IL-4 and IL-8, both showed higher concentrations in the T_FR than the T_b (Table 3, Fig. 3). The median concentration of IL-4 at 8h was 3.1 ng/mL in the T_FR and 2.2 ng/mL in the T_b. At the 4d visit, T_FR concentration was 3.5 ng/mL and T_b concentration was 0.9 ng/mL. IL-8 concentrations were lower in general at the 8h (0.4 ng/mL in the T_FR and 0.1 ng/mL in the T_b) and 4d visits (0.2 ng/mL in the T_FR and 0.0 ng/mL in the T_b). There were no statistical differences found between these analytes (Table 3).

Table 3.

Tear Cytokines and MMPs. Concentrations of IL-4, IL-8, MMP-7. -9 and -10 in each of the 5 tear samples collected from 10 subjects, shown as ng/mL. Significant differences were seen between the T_b and T_FR in MMP-9 after 8h SL wear, and in MMP-10 after 8h and 4d wear. No other significant differences were observed between sample types.

Analytea	Pre-SL	8h SL wear			4d SL wear
	Tb	Tb	TFR	p-value	Tb	TFR	p-value
IL-4	3.7 (0.7; 6.6)	2.2 (0.5; 5.6)	3.1 (0.8; 10.5)	0.20	0.9 (0; 9.3)	3.5 (0.9; 12.3)	0.09
IL-8	0.2 (0; 2.6)	0.1 (0; 0.8)	0.4(0; 3.1)	0.40	0.0 (0; 0.7)	0.2 (0; 1.6)	> 0.99
MMP-7	50.7(30.4; 132.7)	46.0(8.1; 126.0)	54.4(16.5; 183.8)	> 0.99	34.2(8.9; 104.9)	87.5(27.7; 240.7)	> 0.99
MMP-9	31.5(0; 94.4)	15.2(0; 85.1)	62.7(13.7; 300.7)	0.047 *	0(0; 10.5)	18.4(5.7; 86.1)	0.24
MMP-10	13.0(1.3; 18.7)	2.8(0.6; 8.8)	25.8(6.8; 45.2)	< 0.001 **	2.1(0.7; 4.3)	17.2(2.8; 55.1)	0.047 *

^aconcentration shown as median ng/mL (interquartile range).

^*significant p-value comparing the T_b to the T_FR using Dunn’s multiple comparisons test (p < 0.05).

^**significant p-value (p < 0.01).

T_b: basal ocular surface tears. T_FR: fluid reservoir tears.

Fig. 3.

Changes in IL-4 and IL-8 levels traced for each subject for the 5 tear samples collected from each subject at baseline, after 8h and after 4d of SL. T_b: basal ocular surface tears; T_FR: fluid reservoir tears.

In addition, the number of subjects who showed more than a 2-fold greater concentration in the T_FR were counted. After both 8h and 4d of SL wear, the concentration of IL-4 was > 2-fold more in the T_FR in 6 out of 10 subjects. When comparing IL-8 in the T_b and T_FR, 4 of 10 subjects had > 2-fold IL-8 in the T_FR after 8h, and only 3 out of 10 after 4d. While not all subjects showed this magnitude of increase in the T_FR, the remaining subjects showed similar or scattered levels of analytes in the T_b and T_FR, and there were no trends toward greater concentrations of IL-4 or IL-8 in the T_b (Fig. 2).

Fig. 2.

Pie charts showing the percent of subjects with greater than 2-fold concentration in the T_FR compared to the T_b, represented in dark grey for each analyte at both timepoints. Non-shaded areas represent all other subjects that showed more similar concentrations, or greater concentration in the T_b.

Due to the association of MMPs with ocular surface epithelial defects and inflammation, we examined the concentration of MMPs -7, -9 and -10 (Table 3, Fig. 4). There were no significant differences in MMP-7 levels at any of the samples. MMP-9 and -10 were both greater in the T_FR samples, showing a significant difference at 8h (p-value = 0.047 and p < 0.001, respectively), and for MMP-10 only at 4d (p-value = 0.047) (Fig. 4). MMP-9 levels were > 2x higher in the T_FR than the T_b in 8 out of 10 subjects at the 8h and 4d visits. MMP-10 trends were similar, and after 8h SL wear the concentration was greater in the T_FR in 10 out of 10 subjects, and after 4d SL wear 9 out of 10 subjects had > 2x concentration of MMP-10 in the T_FR (Fig. 2).

Fig. 4.

Changes in MMP-7, -9 and -10 levels traced for each subject for the 5 tear samples collected from each subject at baseline, after 8h and after 4d of SL. T_b: basal ocular surface tears; T_FR: fluid reservoir tears; *p ≤ 0.05, **p <0.01.

TIMPs were only tested in selected samples when remaining volume after cytokines/MMPs analysis was available. Due to a lack of entire sample sets for TIMP data, it is not included in the analysis. The range of the MMP-9 and MMP-10 ratios that were calculated with TIMP-1 and TIMP-2 concentrations did not show any patterns and were relatively consistent across all samples except the Day 4 basal tears, in which the ratios were lowest. No conclusions can be made about the TIMP data in this study, but this should be tested in future studies to show the inhibition of MMPs during SL wear.

3.4. Comfort & ocular health evaluation

To evaluate basic satisfaction of neophytes following SL wear, comfort and vision data were analyzed for the 10 subjects that underwent tear analysis. There were no differences between the average CLDEQ score measured pre-SL (11 ±2) compared to after 8h (10 ± 2) or 4d (14 ± 2) of SL wear (p = 0.19) (Table 4). The average VAS score prior to SL was 80.35 ± 6.99 out of the 100-point scale. After 8h of SL wear the VAS comfort was 65.79 ± 6.42 and after 4d it was 60.72 ± 7.87, reduced but not significantly (p = 0.09).

Table 4.

Comfort, staining, and visual acuity at baseline and after 8h and 4d SL wear. Staining scores are a modified NEI grading scale (conjunctiva) – scoring each of 4 quadrants on the 1–3 scale and adding them for a total possible score of 12; and a modified Oxford grading scale (cornea) – scored in 5 corneal areas on a 1–5 scale for a total possible score of 25.

	CLDEQ-8	VAS	Cornea(NaFl)	Conjunctiva (Lissamine)	Visual Acuity (logMAR)
Pre-SL	11 ± 2	80.35 ± 6.99	1.6 ± 0.9	3.0 ± 0.5	−0.09 ± 0.01
8h SL	11 ± 2	65.79 ± 6.42a	1.6 ± 0.5	4.1 ± 0.6	−0.07 ± 0.01
4d SL	14 ± 2	60.72 ± 7.87a	1.4 ± 0.6	3.3 ± 0.7	−0.11 ± 0.02
p-value	0.19	0.09	0.72	0.24	0.16

^asignificant difference compared to pre-SL wear using ANOVA with Dunnett’s post hoc multiple comparison test (≤0.05). Data is shown as mean ± SE.

Change in comfort score was also calculated, positive values indicating improved comfort after SL wear. On average, change in CLDEQ comfort was +2 ±2 after 8h of wear (range −11 to +8) and −3 ± 3 after 4d of SL wear (range −13 to +12). Change in VAS comfort score was −14.56 ±9.13 after 8h of wear (range −63.15 to +38.95) and −19.63 ± 10.68 after 4d of wear (range −71.35 to +37.61). The CLDEQ and VAS scores were compared to each other for each subject, to test correlation of the two testing methods. The inversely-related scoring systems were not strongly correlated at baseline (r = −0.573, p = 0.08) but did have significant correlation at 8h (r = −0.752, p = 0.01) and 4d post-SL wear (r = −0.744, p = 0.01).

There was no change in vision with SL wear compared to spectacle acuity measured at the SL dispense (p = 0.16). Ocular surface staining was evaluated for all subjects at each time point. Conjunctival staining (total potential score of 12) did not change from baseline (3.0 ± 0.5) to 8h (4.1 ± 0.6) or 4d (3.3 ± 0.7) of SL wear (p = 0.24). Corneal staining (total potential score of 25) was 1.6 ± 0.9 pre-SL wear, 1.6 ± 0.5 after 8h, and 1.4 ± 0.6 after 4d of SL wear (p = 0.72). No subjects showed a single sector staining score of greater than 2 for corneal or conjunctival staining, indicating that there was no severe staining associated with SL wear in this study.

There were no relationships between comfort, visual acuity or staining with the levels of the tear analytes tested in this study. For example, of the 6 subjects with the greatest increase of MMP-9 in the T_FR, 2 of them reported improved comfort, 2 of them reported worse comfort, and 2 of them didn’t report a clear change in comfort at all. For MMP-10, the 9 subjects with over 2x greater concentration in the T_FR at 4d also showed no clear relationship between improved comfort (n = 2), worse comfort (n = 4), or no change in comfort (n = 3). Similar trends were seen for vision and corneal/conjunctival staining scores when they were compared to the tear analytes.

4. Discussion

Ocular surface inflammation in SL wearers is a growing concern and understanding the relationship between SLs and the inflammatory state of the eye is essential. This is the first study to determine the differences between the FR microenvironment and the local basal tears. The results show that after 8h and 4d of SL wear there are greater concentrations of MMP-9 and -10 in the FR, and that the increase is dampened after 4 days. The SL were fitted on normal eyes to collect pilot and control data, following standard guidelines to avoid inflammation due to a poor fit.

There were no trends in fitting characteristics (e.g. apical clearance, landing zone appearance) that indicated a risk of increased inflammation with a specific fit, although this study was not designed to test this. The peripheral fit of the SL was designed to reduce excessive conjunctival compression, with 35 % of SL manufactured with toric peripheral curves to accommodate uneven scleral curvature. Lenses had adequate apical clearance to avoid mechanical interaction with the cornea. A study using variable SL fits (e.g. 200 vs. 600 μm apical clearance, different lens diameters) would show whether certain fitting relationships create more or less inflammation in the FR [34]. Unfortunately, corneal thickness and topographical analysis were not evaluated during the present study to assess for hypoxia, but studies have shown that SL create mild, subclinical hypoxia in normal eyes [36,37].

IL-4 and IL-8 were evaluated due to their role in inflammatory eye conditions. IL-4 is associated with angiogenesis and allergies, increased in the presence of contact lens related allergies such as giant papillary conjunctivitis [47]. IL-8, secreted by epithelial cells and inflammatory cells [42,48], can be elevated during soft contact lens wear [49]. No significant changes in IL-4 or IL-8 were observed after SL wear, and no corneal infiltrates were observed. The trend of greater IL-4 in the FR is justification for future studies to look at this marker in a larger cohort and in diseased eyes.

Due to the implications of MMP-9 in dry eye, inflammation, and reduced epithelial barrier function, it is a commonly used as inflammatory marker in the tears [45,50]. In the present study, MMP-9 levels were elevated in FR samples collected after SL wear for 8h and 4d, although only significantly after 8h. Compared to a clinical threshold, 40 ng/mL as used in the InflammaDry^® test to indicate clinically significant inflammation [51], MMP-9 levels are greater only after 8h of SL wear in the FR (median: 62.7 ng/mL). The wide range of MMP-9 was greatest in the FR, which was as high as 659 ng/mL after 8h and > 1000 ng/mL after 4d of SL wear. It should be noted that levels do normally increase into those ranges overnight, as have been measured immediately upon awakening by Markoulli et al. [52], who also measured midday concentrations of 9.8 ± 14.2 ng/mL in the same cohort. The concentration of MMP-9 and the diurnal variations in the FR should be studied further, specifically in diseased eyes which may tend to produce more of this potentially damaging analyte.

MMP-10 is not as well studied in the tear film or cornea, compared to MMP-9. The protease is implicated in wound healing and tissue remodeling, and can be elevated after corneal surgery in diabetic patients [53] and during desiccating corneal stress [41]. Several studies in other tissues have suggested a regulatory role of MMP-10 which may contribute to reducing excessive and potentially damaging effects of inflammation [54]. Here, MMP-10 was markedly elevated in the FR, and its presence may represent a response to SL wear in order to regulate inflammation in the FR. Again, these findings merit the need for future studies to explore the specific origin and implication of increased MMP-10 in the FR.

The increased MMP-9 and -10 in the FR may be due to trapped fluid on the ocular surface when the SL is applied. Alternatively, there may be increased production of MMP-9 and MMP-10 by the corneal epithelial cells during SL wear, which can accumulate in the FR. Both hypotheses imply that there is minimal tear exchange within the subjects fitted in this study, which is consistent with other published reports of SL wear [22,29]. The FR concentrations were often more similar to the early morning tears, which may harbor an increased inflammatory load due to the closed eye overnight environment [52,55], although subjects were awake for at least 45–60 min prior to morning tear collection. Measurement of the protease activity of MMP-9 in FR by the use of gel zymography to determine the levels of MMP-9 inactive and active forms might provide further functional information but could not be accomplished in the present study due to limited sample amount.

The concentration of tissue inhibitors of MMP (TIMPs) are important to consider when assessing the impact of increased MMP in the tears. As the MMP/TIMP ratio increases, it can be indicative of increasing inflammatory state [56]. In the present study, only limited TIMP data was available and most MMP/TIMP ratios could not be calculated. The ratios that were calculated were less than 0.5, which is not particularly indicative of an inflammatory state; however the limited available data does not allow for any conclusive interpretation. Larger studies including more subjects and simultaneous quantification of MMPs and TIMPs are needed to draw conclusions about TIMP regulation of MMPs during SL wear.

This study shows that comfort was reduced after 4d of SL wear in 53 % and 80 % of normal subjects, according to the CLDEQ and VAS, respectably. Contact lens discomfort is the primary reason for discontinuation of contact lens wear in the US [43], and symptoms are usually quite similar to that of dry eye disease, which is known to have an inflammatory component [57]. However, conversely to dry eye, the contribution of inflammation to the discomfort response experienced by contact lens wearers remains unclear. In this study, no correlation was observed between SL discomfort and inflammatory mediator levels in tears. Comfort was highly variable with SL, somewhat contradictory to reports of improved comfort with SL in diseased eyes [12]. This is likely explained by relative responses; individuals with diseased eyes naturally compare the SL to classically uncomfortable alternatives they’ve become adapted to (e.g. corneal GPs), whereas here the subjective comfort rating was coming from individuals accustomed to soft contact lenses which are typically more comfortable. Care should be taken to accurately state the comfort of the SL in light of the population they are intended for and the alternatives available to that population.

There are several limitations of this study, which was the first to compare several different types of tear samples during SL wear. First, the sample size was small, in part due to challenges of collecting the samples. Sample size analysis using this data recommends 20 subjects for future studies, or 40 if there is a normal group compared to a diseased test group. All subjects had normal eyes, and would not typically be fitted into a SL. However, data derived from normal individuals allows for controlled normative data to be determined. This data will be used for development of a similar study in diseased eyes (e.g. keratoconus) who present high variability and a wide range of abnormalities in their tear fluid. There was also not a good baseline to compare without SL wear, since the only pre-SL measurements were taken in the morning which are subject to diurnal variation. However, this study was not designed to compare the tears pre and post-SL, rather was comparing the tears beneath and outside the SL. Tears were also pooled between eyes, which was necessary due to the risk of not always having enough FR volume beneath each lens. Since these were normal eyes, it was assumed that there was not a significant difference in response between the eyes, but future studies may wish to keep eyes separate especially if incorporating lens fit into the analysis of the tear response to SL wear.

5. Conclusion

The results from this study found MMP-9 and MMP-10 were increased in the FR when compared to the basal tears on the ocular surface outside the SL margin. These results suggest inflammatory mediators can become trapped in the FR, which could compromise the ocular surface integrity with prolonged wear. This scenario could be amplified if placed on a diseased eye where MMPs and inflammatory mediators could become chronically trapped in the FR. Therefore it is imperative that future studies continue to evaluate inflammation in the FR during SL wear, in both normal and diseased eyes.