Safety and efficacy of riboflavin-assisted collagen cross-linking of cornea in progressive keratoconus patients: A prospective study in North East India.

INTRODUCTION: Riboflavin- and ultraviolet (UV)-A-mediated collagen cross-linking of the cornea is a frequently used therapeutic measure for the treatment of progressive keratoconus (PK). First, riboflavin increases cross-linking and second, it serves as a protective shield to other deep ocular structures. However, pharmacogenomic variation in riboflavin efficacy is reported. As the Northeast Indian population represents a genetically diverse group of population compared to mainstream India, we have assessed the efficacy of the procedure in a northeastern population with PK.
METHODS: In this study, 78 eyes with PK were included (n = 39 in the treatment arm and n = 39 in the control arm). The primary objective was to evaluate the effect of corneal collagen cross-linking using riboflavin (C3R) (epithelium off) on maximum keratometry. The secondary objectives were evaluation of change in corneal topography parameters, i.e., minimum keratometry (Kmin), simulated keratometry (Sim K), subjective refraction (cylinder power, spherical power, and spherical equivalent [SE]), uncorrected visual acuity (UCVA), best-corrected visual acuity (BCVA), and contrast sensitivity (CS) and safety (intraocular pressure, endothelial cell density, and percentage hexagonality) at 1, 3, and 6 months following C3R procedure.
RESULTS: Statistically significant improvement was noted in Kmin (6 months), Sim K (3 and 6 months), cylinder power (3 and 6 months), spherical power (3 and 6 months), SE (3 and 6 months), BCVA (6 months), and UCVA (1, 6 months) in the C3R group (n = 39) when compared to the control group (n = 39). The mean CS decreased at 1 month and gradually improved to achieve statistically significant value at 6 months in the C3R group (P < 0.05).
CONCLUSION: Riboflavin-assisted C3R treatment showed promising efficacy in the treatment of PK patients in our population. As the collagen turnover rate of cornea is 2-3 years and the progression of PK is highly variable, we need long-term studies to evaluate the efficacy of C3R over time, requirement of repeat therapy, and safety of repeat cross-linking.

Introduction

“Keratoconus” is a progressive, noninflammatory, ectatic condition of the cornea (commonly bilateral). It is characterized by change in the structural organization of corneal collagen fibrils.[1] The clinical hallmarks of the disease are thinning of corneal stroma and subsequent cone-like protrusion, commonly manifesting as refractive astigmatism.[1] In the later phase, the Descemet's membrane may break leading to acute corneal hydrops, causing significant visual impairment.[2] The diagnosis of keratoconus is made based on the presence of two of the following three mentioned criteria: anterior or posterior corneal surface steepening, progressively increasing corneal stromal thinning, or a corneal thickness change in the thinnest location.[3] If disease progression occurs significantly over a period of last 1 year (as evidenced by topography, pachymetry, and refraction), the disease is said to be progressive.[4]

Ultraviolet (UV)-A assisted collagen cross-linking using riboflavin (C3R) is a commonly used therapeutic measure in the management of progressive keratoconus (PK).[4567] In collagen cross-linking process, riboflavin performs multiple roles. It acts as a photosensitizer and induces physical cross-linking between collagen fibrils by generation of oxygen free radicals. It also absorbs UV-A and thus prevents UV-A exposure and subsequent damage of deeper ocular structures, and thus decreases risk to the lens and retina by providing a “shield effect.”[8] In rabbit eye, compared to epithelial off + UV-A alone, higher elasticity was seen in the epithelial off + UV-A + riboflavin-treated rabbit corneas as evidenced from increase in maximal stress and Young's modulus.[2] Here comes the importance of riboflavin in collagen cross-linking.[2] Absorption of riboflavin applied topically is hindered because of corneal epithelial tight junctions. Hence, for optimization of the penetration of riboflavin, the current gold standard treatment protocols involve de-epithelialization, which enhances the absorption of riboflavin.[9] However, response to riboflavin is subjected to pharmacogenomic variation,[10] and variation in riboflavin efficacy is seen in different settings.[11]

In clinical settings, corneal collagen cross-linking treatment shows improvement in vision,[679121314] improvement in apical keratometry,[461516] contrast sensitivity (CS),[1718] a reduction in corneal curvature,[461516] spherical equivalent (SE) refraction,[461415] and refractive cylinder,[461415] with minimal effect on endothelial cell morphology.[1519] However, most of the studies have small sample size, so generalizability of the findings to the whole population is a matter of question. Many of the studies do not have a suitable control group for comparison and to normalize factors associated with the natural history of disease progression. Change in CS (visual ability in distinguishing an object from the background) following C3R procedure is evaluated by only a limited number of studies. Again, there are very few studies showing the effectiveness of collagen cross-linking from India,[2021] and no study has reported the efficacy of the procedure from Northeast India (the northeastern population are geographically different and their genetic makeup is different from the rest of India). With this background of very few reported studies from India and no study from Northeast India, and due to pharmacogenetic variation in riboflavin response, we had undertaken the present study and tried to evaluate visual and morphological changes following corneal collagen cross-linking treatment in PK patients in Northeast Indian population with special reference to change in CS. The aim of the study was to evaluate changes in functional visual quality after corneal collagen cross-linking (epithelial off procedure) for PK patients. The primary objective was to evaluate C3R on maximum keratometry (Kmax) at 6 months. The secondary objectives were evaluation of change in corneal topography parameters (minimum keratometry [Kmin] and simulated keratometry [Sim K]), refraction (spherical power, cylindrical power, and SE), uncorrected visual acuity (UCVA), best-corrected visual acuity (BCVA), and CS and safety (intraocular pressure [IOP], endothelial cell count, and percentage hexagonality) at 1, 3, and 6 months following C3R (epithelial off procedure).

Methods

Study design

This hospital-based prospective study (parallel group, allocation ratio 1:1) was conducted in Sri Sankaradeva Nethralaya, Guwahati, Assam, on 78 eyes of 39 patients with bilateral PK.

Eligibility criteria

For the present study, PK was defined based on a previous report by Wittig-Silva et al.[16] Presence of any one of the following criteria is required for the diagnosis of PK: >1.0 D increase in steepest Sim K, >1.0 D increase in refractive astigmatism, and an increase of >0.1-mm decrease in the back optic zone radius of the best-fitting contact lens in the last 1 year. The staging of keratoconus was based on the Amsler–Krumeich keratoconus classification.[22]

Patients with bilateral PK aged between 15 and 50 years, willing to give written informed consent, with more than 400 μm of minimum central corneal thickness (CCT) with topographic documentation of PK, and with no previous history of ocular surgery were included in the present study. We excluded patients who refused to give consent, those with extreme ages, i.e., <15 years or >50 years of age, those with maximal keratometry readings (Kmax) >60D, CCT of <400 μm, ocular surface disorders, apical corneal scarring, previous ocular surgery, history or clinical sign of hydrops, history of ocular herpes or nonhealing corneal ulcers, previous history of thermal or chemical burn, any associated autoimmune diseases, associated diabetes, and pregnant and lactating mothers.

Preoperative evaluation

A detailed history of the patients was taken regarding onset, symptoms, allergic history, and previous treatment history.

Randomization

In patients with bilateral PK, eyes were randomized by using coin toss method, with the right eye assigned as head and the left eye assigned as tail. According to the toss of coin, one eye served as the treatment eye, and the other eye of the same participant served as the control eye.

Intervention

Control group

Control eyes were given artificial tear drops.

C3R group

Eyes were subjected to riboflavin-assisted, UV-A-mediated collagen cross-linking.

C3R (epithelial off) procedure

Epithelium-off technique (Dresden protocol) as described by Wollensak et al.[6] was followed. Under sterile operating condition, two drops of pilocarpine (2%) were applied (for pupillary constriction and minimizing UV [A] exposure to the lens and retina). The patient was placed in supine position, and the cornea was positioned perpendicular to the aiming laser beam. Proparacaine (0.5%) eye drop was applied as surface anesthesia, and the skin was prepared with 5% povidone-iodine solution. At the central cornea, a circular area (diameter 8 mm) was marked using a corneal trephine. On applying 20% ethyl alcohol, the corneal epithelium swelled which was scrapped off with a hockey stick knife. After a thorough wash of the corneal surface, 0.1% riboflavin (prepared in 20% dextran) was applied for 30 min at 5-min interval. Yellow flare in the anterior chamber (observed by slit-lamp examination) was taken as a marker for the confirmation of riboflavin penetration. UV-A irradiation (365 nm for 30 min) was then performed from 50-mm distance (from the cornea) with an irradiance of 3 mW/cm2 using a UV-X lamp (model-UV-X-1000, PESCHSE, Germany). Riboflavin instillation was continued at 5-min interval throughout the process of UV-A irradiation. After completion of the irradiation procedure, two drops of topical gentamycin were applied, followed by the application of a soft bandage contact lens. The patients were prescribed dark goggles for sun protection and protective eye shields at bedtime. Following the procedure, the patients were prescribed a topical antibiotic and steroid fixed dose preparation for 3 weeks followed by gradual tapering with anti-inflammatory eye drops (nonsteroidal anti-inflammatory drug) for 2 weeks and a preservative-free artificial tear drop for 6 months. The patients were instructed to avoid driving and eye rubbing for 2 weeks.

Follow-up

Postoperative examinations were done on day 1, day 3, and day 7 (vision and slit-lamp examination) till epithelial defect was healed which was followed by repeat visits at the 1st, 3rd, and 6th months.

Outcome evaluation

The outcome evaluator (Pentacam®, Topographic Modeling System [TMS]-4 and specular microscopy evaluator) was blinded to study group allocations. Data collected were corneal topography parameters (Kmax, Kmin, and Sim K), subjective refraction (spherical power, cylinder power, and SE), visual acuity (UCVA and BCVA), CS, IOP, and specular microscopy parameters (endothelial cell count and percentage hexagonality). Data were recorded at the 0, 1st, 3rd, and 6th months. UCVA and BCVA were recorded by using logarithm of the minimum angle of resolution (log MAR) chart at 6 m distance, Pelli–Robson Contrast Sensitivity chart was used for the measurement of CS (at 4-m distance), Kmax and Kmin were measured by Pentacam (OCULUS, Pentacam HR, Germany), Sim K was measured by TMS-4 (TMS 4 Topography modelling system, TOMEY, Germany), endothelial cell count and percentage hexagonality were measured by specular microscopy (SP 2000, Topcon, Japan), and IOP measurement was done using Goldman applanation tonometer (AT 900, model R, HAAG-STREIT diagnostics, Switzerland).

For topographical evaluation, prior discontinuation of contact lens was advised (3 days for soft contact lens and 3 weeks for hard contact lens). A stable refraction was necessary in two consecutive examinations, with a minimum of 1-week interval for contact lens wearers before enrolling into the study. (The difference of mean refractive SE and keratometry should not be more than 0.75 D).

Ethical considerations

The study was conducted after approval from the institution's Ethics Committee and Scientific and Research Committee (EC no. 19, dated March 12, 2015, project no. 5). Before enrollment, informed consent was taken from all the participants. Clinical trials registry India (CTRI) reference number was REF/2018/05/020083.

Sample size calculation

Using the Kmax values in the treated and untreated groups as 53.82 ± 0.72 D and 53.31 ± 0.72 D, respectively,[23]95% confidence interval (two sided), 80% power, and case-to-control ratio of 1:1, a sample size of 64 was calculated (32 in each group). Taking a 20% dropout, a sample size of 78 (39 in each group) was calculated.

Statistical analysis

The collected data were subjected to a master tabulation in MS Excel spreadsheet. Data were analyzed by using IBM SPSS Statistics for Windows, Version 22.0. (IBM Corp., Armonk, NY). Depending on the distribution of data, suitable statistical tests for hypothesis testing were used along with post hoc tests. Repeated-measures ANOVA was used to analyze the repeated-measures data collected at different time points (follow-ups). P < 0.05 was termed statistically significant.

Results and Observation

Participant flowchart is shown in Figure 1. Out of 68 bilateral keratoconus patients screened, 39 participants met the inclusion/exclusion criteria and gave written informed consent for participation in the study. Various outcomes of the procedure are summarized in Table 1, and the adverse effect data are shown in Table 2. Pentacam image of representative cases (C3R-treated patients) and controls is depicted in Figure 2.

Figure 1

Participant flowchart

Table 1

Visual and morphological changes following collagen cross-linking

Parameter	Group	Baseline	1 month	P$	3 months	P$	6 months	P$	P#
K Max	Control	58.6±7.2	58.39±7.07	1.00	59.31±7.3	1.00	59.97±7.34	1.00	<0.001
	C3R	59.02±7.3	58.77±7.29	1.00	57.62±7.17	1.00	57.39±7.21	1.00
	p*	0.805	0.817		0.308		0.122
K Min	Control	48.81±4.24	48.82±4.16	1.00	49.18±4.15	1.00	49.54±4.00	1.00	<0.001
	C3R	48.12±4.04	47.95±3.99	1.00	47.66±3.94	1.00	47.13±4.03	1.00
	p*	0.464	0.344		0.101		0.010
Sim K	Control	54.80±4.63	54.97±4.59	1.00	55.71±4.75	1.00	56.12±4.84	1.00	<0.001
	C3R	54.78±4.58	54.15±4.33	1.00	53.33±4.44	0.980	52.71±4.67	0.267
	p*	0.987	0.423		0.025		0.002
Cylinder power	Control	−2.65±0.82	−2.78±0.92	1.00	−3.09±0.98	0.217	−3.39±0.89	0.003	<0.001
	C3R	−2.84±0.85	−2.51±0.86	0.715	−2.16±0.92	0.010	−1.65±1.09
	p*	0.315	0.187		<0.001		<0.001
Spherical power	Control	−2.04±0.39	−2.04±0.39	1.00	−2.51±0.43	<0.001	−3.05±0.60	<0.001	<0.001
	C3R	−2.07±0.44	−2.07±0.44	1.00	−1.34±0.61	<0.001	−0.82±1.02	<0.001
	p*	0.738	0.738		<0.001		<0.001
SE	Control	−3.17±0.67	−3.17±0.67	1.00	−3.95±0.86	<0.001	−4.67±0.83	<0.001	<0.001
	C3R	−3.26±0.84	−3.26±0.84	1.00	−2.43±0.87	0.005	−1.77±1.54	<0.001
	p*	0.606	0.606		<0.001		<0.001
BCVA	Control	0.115±0.097	0.1426±0.0964	1.00	0.2023±0.1103	0.003	0.2638±0.1218	<0.001	<0.001
	C3R	0.1251±0.944	0.2826±0.0891	<0.001	0.2074±0.0758	0.001	0.0674±0.1060	0.038
	p*	0.674	<0.001		0.812		<0.001
UCVA	Control	0.318±0.143	0.3051±0.1316	1.00	0.4026±0.1813	0.109	0.4590±0.1649	0.001	<0.001
	C3R	0.323±0.1224	0.4926±0.2704	0.002	0.3674±0.1691	1.00	0.2774±0.2267	1.00
	p*	0.865	<0.001		0.379		<0.001
CS	Control	1.1538±0.4515	1.1346±0.43228	1.00	0.9769±0.34562	0.270	0.8962±0.29500	0.022	<0.001
	C3R	1.1538±0.45543	0.9846±0.45885	0.687	0.9723±0.49270	0.544	1.1154±0.47533	1.00
	p*	1.00	0.141		0.962		0.017
IOP	Control	12.48±2.94	12.66±2.65	1.00	12.84±2.25	1.00	12.56±2.29	1.00	0.594
	C3R	12.41±2.92	12.79±2.44	1.00	12.76±2.22	1.00	12.23±2.49	1.00
	p*	0.908	0.825		0.880		0.541
Endothelial cell count	Control	2723±527	2836±243	0.859	2831±241	0.958	2822±240	1.00	0.424
	C3R	2747±345	2791±249	1.00	2786±247	1.00	2780±245	1.00
	p*	0.818	0.424		0.420		0.457
Percentage hexagonality	Control	70.25±15.72	67.35±16.38	1.00	64.94±17.39	0.941	62.46±16.35	0.230	0.008
	C3R	73.41±13.90	67.56±14.27	0.474	64.82±15.21	0.062	62.92±14.97	0.011
	p*	0.351	0.953		0.972		0.897

P* When between the two groups at the same time point, P$ when compared to baseline in the same group, P# stands for overall p value of the trend. K Max=Maximum keratometry, K Min=Minimum keratometry, Sim K=Simulated keratometry, BCVA=Best-corrected visual acuity, UCVA=Uncorrected visual acuity, CS=Contrast sensitivity, SE=Spherical equivalent, IOP=Intraocular pressure

Table 2

Adverse events reported in the corneal collagen cross-linking and the control group until 6 months of follow-up

Adverse events	C3R (n=39)	Control (n=39)
Pain and congestion	15 (38.4%) cases (up to 3rd postoperative days), treated with topical NSAIDs	Nil
Glare	24 (61.5%) patients had glare during night in 1 month, which improved spontaneously by 3rd months	Nil
Corneal stromal haze	All patents (100%) had corneal stromal haze till 1 month, which spontaneously reduced by 3rd months	Nil
Superficial punctate keratitis	1 (2.56%) patient (treated with topical antibiotics and artificial tear)	Nil
Epithelial defect	1 (2.56%) patient (treated with topical antibiotics and artificial tear)	Nil
Nebular corneal opacity	5 (12.8%) cases (all cases with nebular corneal opacity used contact lens postoperatively)	Nil
Macular corneal opacity	1 (2.56%) patient (used contact lens postoperatively)	Nil
Disease progression	4 cases	All cases

NSAIDs=Nonsteroidal anti-inflammatory drops, C3R=Corneal collagen cross-linking

Figure 2

Pentacam picture of cases and controls before treatment and after 6 months following treatment

Demographic parameters

In our study, mean age in both the groups was 23.795 ± 5.44 years, and 66.6% of the patients were male and 33.3% were female (the demographic parameters in both the groups were similar as one eye of the same person served as control, whereas the other eye was in the C3R arm).

Corneal topography outcomes

Effect of riboflavin-assisted collagen cross-linking using riboflavin on maximum keratometry value

An increase of 1 D or more in the Kmax reading was considered to be progression of the ectasia. A decrease in Kmax reading more than 1 D was considered to be regression. The ectasia was considered to be stable in those cases where the Kmax value continued to be within ± 1D.

In our study [data shown in Table 1 and Pentacam image shown in Figure 2], no significant decrease in Kmax value was seen in both C3R and control groups in each follow-up (comparing both the groups), but P values showed a decreasing trend (P = 0.817, 0.308, and 0.112 at the 1st, 3rd, and 6th months, respectively). In the control group, the mean Kmax values increased from the baseline values of 58.6 ± 7.2 D to 59.31 ± 7.3 D at 3 months and 59.97 ± 7.34 D at 6 months, but they were not statistically significant. In the C3R group, compared to its baseline value (59.02 ± 7.3 D), treatment resulted in decrease in mean Kmax at the 1st month (58.77 ± 7.29 D), 3rd month (57.62 ± 7.17 D), and 6th month (57.39 ± 7.21 D), but the difference was not significant statistically.

Effect of riboflavin-assisted C3R on minimum keratometry value

Corneal topographical analysis revealed [data shown in Table 1] that mean Kmin value was gradually decreasing from its baseline value (48.12 ± 4.04 D) to the last follow-up period of 6 months (47.13 ± 4.03 D) in the C3R group. The control group showed gradual increase in trend from its baseline (48.81 ± 4.24 D) to that of the last follow-uP value in 6 months (49.54 ± 4.00 D). The difference between both groups was statistically significant in the 6th month postoperatively (P = 0.010).

Effect of riboflavin-assisted collagen cross-linking using riboflavin on simulated keratometry value

In the control group [data shown in Table 1], the mean Sim K value increased from the baseline value of 54.80 ± 4.63 D to 54.97 ± 4.59 D at the postoperative 1st month, 55.71 ± 4.75 D at the 3rd month, and 56.12 ± 4.84 D at the 6th month. On the other hand in the C3R group, Sim K values decreased from the baseline value of 54.78 ± 4.58 D to 54.15 ± 4.33D at the 1st month, 53.33 ± 4.44 D at the 3rd month, and 52.71 ± 4.67 D at the 6th month. The difference between both the groups reached statistically significant level at month 3 and month 6 (P = 0.025 and 0.002, respectively).

Refractive outcomes

Effect of riboflavin-mediated collagen cross-linking using riboflavin on cylinder power

In the control group [data shown in Table 1], the mean cylinder power increased from its baseline value of −2.65 ± 0.82 D to −2.78 ± 0.92 D at the 1st month, −3.09 ± 0.98 D at the 3rd month, and −3.39 ± 0.89 D at the 6th month, with statistically significant increase in cylinder power at the 6th month (P = 0.003), when compared to baseline. While in the C3R group, the mean cylinder power decreased from its baseline value of −2.84 ± 0.85 D to −2.51 ± 0.086 D at the 1st month, −2.16 ± 0.92 D at the 3rd month, and −1.65 ± 1.09 D at the 6th month. The difference between the two groups (C3R and control) reached statistical significance in the 3rd and 6th months (P < 0.001 at both visits in comparison to baseline).

Effect of riboflavin-assisted collagen cross-linking using riboflavin on spherical power

In the control group [data shown in Table 1], the mean spherical power increased from the baseline value of −2.04 ± 0.39 D to −2.08 ± 0.45 D at the 1st month, −2.51 ± 0.43 D at the 3rd month, and −3.05 ± 0.06 D at the 6th month, with statistically significant increase in spherical power at the 3rd and 6th months (P < 0.001 on both the visits). In the C3R group, mean spherical power decreased from the baseline value of −2.07 ± 0.44 D to −2.02 ± 0.45 D at the 1st month, −1.34 ± 0.61 D at the 3rd month, and −0.82 ± 1.02 D at the 6th month, with statistically significant decrease in spherical power at the 3rd and at the 6th months (P < 0.001 in both visits) when compared to the baseline value. The difference between the two groups (C3R and control) reached statistical significance in the 3rd and 6th months (P < 0.001 and P < 0.001, respectively).

Effect of riboflavin-assisted collagen cross-linking using riboflavin on spherical equivalent

In the control group [data shown in Table 1], the mean SE increased from the baseline value of −3.37 ± 0.61 D to −3.47 ± 0.67 D at the 1st month, −4.05 ± 0.73 D at the 3rd month, and −4.7 ± 0.86 D at the 6th months, with statistically significant increase in SE in the 3rd and 6th months (P < 0.001 in both visits), when compared with the baseline value. While in the C3R group, SE decreased from the baseline value of −3.5 ± 0.57 D to −3.28 ± 0.57 D at the 1st month, −2.42 ± 0.86 D at the 3rd month, and −1.64 ± 1.38 D at the 6th month, with statistically significant decrease in SE at the 3rd and 6th months (P < 0.001 in both visits), when compared with the baseline value. The difference between the two groups (C3R and control) reached a statistical significance in the 3rd and 6th months (P < 0.001 in each follow-up).

Visual acuity endpoints

Effect of riboflavin-assisted collagen cross-linking using riboflavin on best-corrected visual acuity

In the control group [Table 1], mean BCVA in log MAR increased from its baseline value of 0.115 ± 0.09 log MAR to 0.14 ± 0.09 log MAR (1st month), 0.20 ± 0.11 log MAR (3rd month), and 0.26 ± 0.12 log MAR (6th month), with statistically significant increase at the 3rd month (P = 0.003) and at 6th month (P < 0.001) when compared to the baseline value. On the other hand, in the C3R group, mean BCVA increased from the baseline value of 0.12 ± 0.94 log MAR to 0.28 ± 0.08 log MAR at the 1st month (P < 0.001). It started decreasing by 0.20 ± 0.07 log MAR at 3 months (P = 0.001) and decreased subsequently to 0.06 ± 0.10 log MAR at 6 months (P < 0.001), when compared to baseline. The difference between the two groups (C3R and control) reached statistical significance at the 1st and 6th months (P < 0.001 and P < 0.001, respectively).

Effect of riboflavin-assisted collagen cross-linking using riboflavin on uncorrected visual acuity

In the control group [data shown in Table 1], mean UCVA increased from the baseline value of 0.318 ± 0.143 log MAR to 0.402 ± 0.181 log MAR at the 3rd month and 0.459 ± 0.164 log MAR at the 6th month, with a statistically significant increase at the 6th month (P < 0.001), when compared to baseline. On the other hand, in the C3R group, UCVA increased from the baseline value of 0.32 ± 0.12 log MAR to 0.49 ± 0.27 log MAR at the 1st month (P = 0.002). It started decreasing by 0.36 ± 0.16 log MAR at the 3rd month (P = 1.00) and then decreased subsequently to 0.06 ± 0.10 log MAR at the 6th month (P = 1.00), when compared to baseline. The difference between the two groups (C3R and control) reached statistical significance at the 1st month and at 6th month (P < 0.001 in both follow-ups).

Effect of riboflavin-assisted collagen cross-linking using riboflavin on contrast sensitivity

In the control group, the mean [data shown in Table 1] CS decreased from the baseline value of 1.15 ± 0.45 log CS to 1.13 ± 0.43 log CS at the 1st month, 0.97 ± 0.34 log CS at the 3rd month, and 0.89 ± 0.29 log CS at the 6th month, with statistically significant decrease at the 6th month (P = 0.022), when compared to baseline. On the other hand, in the C3R group, CS first decreased from the baseline value of 1.15 ± 0.4 log CS to 0.98 ± 0.45 log CS at the 1st month (P = 0.687). It started increasing by 0.97 ± 0.49 log CS at the 3rd month (P = 0.5) and then increased subsequently to 1.11 ± 0.47 D at the 6th month (P = 1.00), when compared to baseline. The difference between the two groups (C3R and control) reached statistical significance at the 6th postoperative month (P < 0.01).

Safety endpoints

Effect of riboflavin-assisted collagen cross-linking using riboflavin on intraocular pressure

There was no significant difference in IOP [referred to Table 1] between both groups in any follow-up when compared with the control group. No significant change was noted in the repeated observations when compared to the respective baseline data.

Effect of riboflavin-assisted collagen cross-linking using riboflavin on endothelial cell density

No significant difference was observed between the two groups [Table 1] at any of the time points, as compared to the control group. No significant change was noted in the repeated observations when compared to the respective baseline data.

Percentage hexagonality

No significant difference was seen between C3R and control groups [data shown in Table 1] in any follow-up period.

Adverse events

The data of all the adverse events reported in both groups are shown in Table 2. Pain, congestion (38.4%), glare (61.5%), and corneal stromal haze (100%) were the most common adverse events reported following the cross-linking procedure. In the control group, except for disease progression, no other adverse event was recorded.

Discussion

Riboflavin-assisted corneal collagen cross-linking is considered one of the most promising treatments in PK nowadays.[1011152021222324252627] The present study reports visual and morphological outcome in 78 eyes over a period of 1 year in a prospective setting of riboflavin-assisted C3R treatment. The results of the current study demonstrate statistically significant improvement in UCVA, BCVA, CS, Kmin, Sim K, spherical and cylinder power, and SE in the treatment group. Improvement in Kmax value was not found to be statistically significant at the maximum follow-up period of 6 months in the treatment group. In the control group, the Kmax, Kmin, and Sim K increased and UCVA, BCVA, CS, spherical power, cylinder power, and SE deteriorated, suggesting continuing progression of the disease. No significant changes were seen in IOP, endothelial cell density, and percentage hexagonality in both the control and C3R groups.

Maximum keratometry value

Kmax is a topographical indicator for the success of riboflavin-assisted C3R as it measures the severity of the disease process.[4] It denotes the steepest anterior corneal curvature, is used to document ectatic progression, and is an indicator for cross-linking efficacy.[4]

Pioneering works by Wollensak et al.[6] and subsequent studies[41226] reported statistically significant decrease in Kmax value following C3R therapy. The reason for decrease in the Kmax value is due to the flattening effect on the cone by the process of cross-linking.[2829] No significant decrease in the Kmax value was noted till 3 months post-C3R therapy.[426] Statistically significant decrease of Kmax value was seen at the 6th month[426] and maintained till the 12th month,[42630]2nd year,[26] and 3rd year,[16] post-C3R therapy. Rearrangement of corneal stromal collagen is hypothesized to play in the whole process.[29] Considering the natural collagen turnover process of cornea, long follow-up studies are required to determine the necessity of repeat C3R treatment. In our study, although the mean values of Kmax were less compared to the control group at 3 and 6 months, they were not statistically significant. Progression was seen in the control group, whereas progression was halted in the C3R group, in our study.

Minimum keratometry value

Kmin value is also a measure of treatment efficacy in collagen cross-linking. In our study, in the treated group, although the Kmin value decreased in all the follow-ups, when compared to baseline data, the difference was not statistically significant till 3 months. At 6 months, significant difference was seen while comparing the treated group with the control group. The control group showed an increasing trend in Kmin value from baseline, which corresponds to disease progression, but this increase in Kmin was not statistically significant, even at 6 months. The Kmin value decreased in the C3R group and when compared to the control group, this difference was statistically significant at 6 months. It seems that change in Kmin, which is an indicator of treatment efficacy, starts gradually. Hersh et al.[4] found no significant difference between the C3R and control groups in Kmin value at the 3rd and 6th months and became significant at 1-year follow-up. Similarly, Wittig-Silva et al.[1626] found significant difference between the C3R and control groups which appeared in 12 months and maintained till the 3rd year. We need long-term studies to observe how long this effect progresses and how long it is maintained.

Simulated keratometry value

Maximum central curvature is measured by Sim K measurement (Sim K) or corneal astigmatism. It gives an idea of corneal curvature of the central visually significant 3-mm area. Sim K value was recorded from the topography generated data as determined by the TMS 4 system. The regression of Sim K is an indicator of treatment efficacy. In our study, in the C3R group, Sim K decreased gradually following C3R treatment and the difference became statistically significant which was observed in the 3rd and 6th months. When compared to baseline data, in the control group, the value of Sim K increased gradually. Similar decrease in Sim K value following C3R therapy was observed by Viswanathan et al.,[24] Vinciguerra et al.,[12] Hersh et al.,[4] and Lamy et al.[17] O'Brart et al. in their preliminary[31] and long-term studies[32] observed that there was a significant decrease in Sim K value in the 1st year and 18th month and maintained till 4–6 years.

Cylinder power

In our study, a trend of decrease in cylinder power was seen in the C3R group. On the contrary, the control group showed continuous progression of refractive astigmatism. Difference in cylinder power between the two arms was significant at the 3rd and 6th postoperative months. Similar decrease in cylinder power was observed by other studies at as early as 2 months,[14] 3 months,[14] 1 year,[61230] and 18 months[31] and was maintained up to 4 years.[32] The flattening effect of the cone due to the cross-linking is the reason for the significant regression of cylinder power.

Spherical power

In the present study, spherical power of the treatment group showed a decreasing trend from its baseline value. On the contrary, in the control group, there was a gradual progression of spherical power from the baseline till 6 months. At the 3rd and 6th months, a statistically significant difference was noted between the two groups. Similar result was documented by both Caporossi et al.[14] and Vinciguerra et al.[12] Caporossi et al.[14] found decrease in the spherical power from −2.85 ± 3.68 D preoperatively to −1.95 ± 2.83 D at the 1st month; −1.875 ± 2.56 D at the 2nd month, and −1.425 ± 2.54 D at the 3rd month. Vinciguerra et al.[12] found significant difference between the treatment and control groups in 1-year follow-up in their study population. On the other hand, Wittig-Silva et al.[1626] found no significant difference among the two groups at any of the follow-ups.

Spherical equivalent

In our study, significant difference in spherical equivalence between both C3R and control arms was observed in the 3rd and 6th months. The control group showed progressive increase of SE from the preoperative value in the 3rd- and 6th-month follow-up period. Similar results were reported by Hersh et al.[4] (significant from baseline at 1, 3, and 6 months), Vinciguerra et al.[12] (at 12 months), Viswanathan et al.[24] (significant from baseline at 20 months), and Wollensak et al.[6] (significant from baseline at 4 years).

Best-corrected visual acuity

In our study, in the C3R group, BCVA dropped in the initial postoperative 1-month period and improved gradually in follow-up periods. In the control group, there was gradual deterioration of vision with a significant difference between the two treatment arms at the 6th month. Similar improvements in BCVA were also noted by Vinciguerra et al.[12] (at 3, 6, and 12 months), Wittig-Silva et al.[16] (1, 2, and 3 years), Viswanathan et al.[24] (20 months), and Raiskup-Wolf et al.[30] (at 1 year). The reason for initial deterioration of vision following C3R in our study could be due to stromal haze induced by the procedure. With subsequent reduction of stromal haze, BCVA improved in the subsequent follow-up period.[17]

Uncorrected visual acuity

Although the primary objective of C3R procedure is to halt the progression of the disease, visual improvement is also seen many a times following the procedure as secondary effect.[41229] In our study, UCVA increased in the initial postoperative 1-month period in the C3R group. With subsequent follow-up, it gradually improved in the 3rd and 6th months and was statistically significant at 6 months between the two groups. In the control group, there was gradual deterioration of vision in the subsequent follow-ups. The reason for initial deterioration of vision is due to stromal haze induced by the procedure.[17] With subsequent reduction of stromal haze, UCVA improved in the subsequent follow-up period. Similar observations were also made by Caporossi et al.[14] (at 2 and 3 months), Vinciguerra et al.[12] (at 3, 6, and 12 months), Hersh et al.[4] (at 12 months), and Wittig-Silva et al.[16] (1, 2, and 3 years). Wittig-Silva et al.[16] noted significant improvement in UCVA of − 0.15 log MAR which was seen at 36 months post-C3R. In the control group, deterioration of UCVA was found in each follow-up period and statistically significant only at the 36th month of follow-up.

Contrast sensitivity

In our study, CS decreased gradually in the control group, whereas in case of C3R, it first deteriorated at the 1st month and then came back to baseline level at 6 months. Statistically significant difference was noted in the 6th month, with significantly higher value in the C3R group. Similar observation was made by Lamy et al.[17] in their study in 68 keratoconic eyes. They observed that the mean CS improved from baseline to 2-year follow-uP value (P < 0.001) in the C3R group. The authors concluded that anterior stromal haze induced by the procedure in the early postoperative period is responsible for diminished CS which improved following clearing of stromal edema.[1830]

Intraocular pressure

In our study, no significant difference was seen between the two groups at any of the follow-ups. Similar results were also found by various literatures.[461214] Wittig-Silva et al.[1626] reported that there was no significant difference in IOP observed in all postoperative follow-ups till 36 months between the groups using Tono-Pen® and Goldmann applanation tonometer.

Endothelial cell density and percentage hexagonality

In our study, no statistically significant change was seen in endothelial cell count and percentage hexagonality in any of the follow-up periods between the two groups (C3R and control) when compared to the respective baseline values. Similar results were also found by Vinciguerra et al.,[12] Holopainen and Krootila,[19] and Wittig-Silva et al.[1626]

Adverse effect

In the present study, in the C3R group, 15 patients complained of pain and congestion up to the 3rd postoperative day, which were treated with topical nonsteroidal anti-inflammatory drops. Twenty-four patients had glare during night in the 1st month which improved spontaneously by the 3rd month. Till 1 month, there was corneal stromal haze in all the patients which subsequently reduced by the 3rd month. One case of superficial punctate keratitis and one case of epithelial defect were reported which were treated by the application of topical antibiotics and lubricating eye drops. By the 6-month postoperative period, six patients with a history of contact lens use developed nebular corneal opacity (five cases) and macular corneal opacity (one case), thus indicating contact lens users to be at risk for developing corneal opacity, however, the developed corneal opacities were not visually significant. Four cases with reduction in visual acuity, increase in cylindrical power by 1 D, and topographical evidence of progression were reported at 6-month follow-up. Those patients were advised for regular follow-up. Similar observations with spontaneous clearing of stromal haze by the 3rd month were reported by various authors.[10182023333435] Caporossi et al.[14] concluded that patients with advanced stage of keratoconus are at an increased risk of corneal haze which develops after C3R, with the causes being low corneal thickness and high corneal curvature.

Conclusion

At present, keratoconus is not a curable disease, and the progression of the disease is highly variable among different patients. However, riboflavin-assisted cross-linking has shown efficacy in stopping the disease progression. In our study, although no significant improvement was noted in the parameters such as Kmax, IOP, endothelial cell density, and percentage hexagonality, other parameters, for example, BCVA, UCVA, CS, Kmin, Sim K, SE, cylinder power, and spherical power significantly improved following C3R treatment at 6 months. The durability of the stiffening effect of C3R is unknown. As the turnover of corneal collagen ranges from 2 to 3 years,[1] long-term data are needed regarding the safety and efficacy of the procedure and to gather data regarding requirement of a repeat procedure. This study highlights the importance of riboflavin-assisted corneal C3R procedure in the treatment of PK.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.