Corneal cross-linking (CCL) was approved in 2016 by the Food and Drug Administration (FDA) as a drug and device combination for the treatment of progressive keratoconus and corneal ectasia after refractive surgery. CCL procedure strengthens the cornea with the purpose of preventing the progression of keratoconus or post-refractive corneal ectasia. In addition, some studies have shown that CCL may also lead to improvement of topographic and visual outcomes in this patient (18).
Most studies in keratoconus have compared the results of different CCL methods such as standard corneal cross-linking or epithelium-off CCL (epi-off CCL) and trans-epithelial corneal cross-linking (TE-CCL) (19, 20, 21). In recent years, several clinical trials have compared the therapeutic effects between the two methods (19, 22, 23), but studies on accelerated CCL are very limited.
Our study aim was to evaluate tomographic, densitometric and aberrometric parameters in patients with progressive keratoconus after accelerated CCL and, due to the fact that appearance and natural course of postoperative haze after corneal cross-linking will assist physicians in knowing what to anticipate postoperatively and in counseling patients regarding postoperative expectations, firstly, we classify densitometric outcomes of accelerated CCL in different stages of the KCN separately (Mild, moderate and severe), and we have answered the question whether the severity of keratoconus affects the corneal haze after CCL?
Accelerated CCL resulted in stabilization of the CDVA during the 12-month follow-up, which is similar to the results obtained in a number of other studies (24, 25), although most of the comparative studies have reported improvement in CDVA. Although our results demonstrated minimal changes in refractive and keratometric values, CDVA were significantly improved, respectively. This visual improvement after accelerated CCL could be explained by the improvement in the Root mean Square (high order aberration).
The two most significant indicators of visual acuity improvement after corneal cross-linking were the low CDVA (≤ 20/40) preoperatively and high maximum keratometry values (≥ 55 D). In the current study, final CDVA at month 12 had correlation with month 3 middle 0–2 mm zone corneal densitometry.
In our study, we found statistically significant flattening in maximum keratometry value after a one year follow-up. In addition, the flattening effect on maximum and mean keratometry were in agreement with the results of previous well-known studies (13, 24, 25). Current study results are consistent with those of recent studies that have decreased Thinnest pachymetry values after accelerated CCL (21, 27). The physiology of this corneal thinning after CCL still not identified: however, structural changes occur in corneal collagen fibrils (28), such as changes in corneal hydration (29), compression of collagen fibrils (30), edema (31), and keratocyte apoptosis (31) are discussed in the literature. It should be borne in mind that previous studies have shown that due to the difference between the Pentacam and ultrasound pachymetry, pachymetric values obtained from Pentacam printout should be interpreted with extreme discretion (32).
Reported in our study, as in the epidemiological study (33), progressive KCN was more common in men than in women.
Root mean Square (high order aberration) is part of the refractive errors which are not correctable with sphere and cylinder corrections, and it is among errors of the optical system of the eye which can exacerbate the quality of the retinal image (34, 35). Since RMS (HOA) can has a considerable effect on the visual function and contrast sensitivity; It’s considered an important value in the context of vision quality (35, 36). In the present study High order RMS value was significantly increased at 3 months after CCL and then decreased at one year, relative to the baseline. Naderan et al(38), and Ahmed et al(39), reported the efficacy of corneal cross-linking in improving HOA parameters in eyes with progressive KCN, and their results showed that there is a statistically significant correlations between pre-operative values of HOAs parameters with corrected distance visual acuity.
In order to determine the prognostic strength of peak densitometric values on final visual outcomes, Correlation between peak densitometric numbers at 3 month post-op and final visual acuity, total RMS and higher order RMS at 12th month were evaluated, and the results showed that RMS at 1 year had correlation with densitometric values of Anterior 0–2 mm area, Anterior 2–6 mm area and total corneal 0–2 mm and 2–6 mm areas at third month. This relationship means that the aberrometric effect in final visual acuity is due to its effect on densitometric factors.
Mathews et al (40) showed for the first time a correlation between densitometric value changes at 6 month and higher order aberrations after CCL with Dresden protocol. We found a similar correlation in eyes with keratoconus after accelerated CCL.
Also, for the first time we studied densitometric factors in different stages of keratoconus patients, our results showed that relationship between Higher order RMS at 12th months and peak densitometric values has different nature in different stages of progressive KCN.
For example, densitometric values on the severe stage had no correlation with RMS at 12th months. But we showed correlation in multiple zones in mild and moderate stage. These results may bring in to mind that in addition to the possible effect of densitometric factor on aberrometric values, there must be other factors that affect aberrometric factors and consequently the patient's CDVA in more advanced stages of KCN and the prognostic value of peak densitometrc parameters is weekend in more advanced stage of the disease, while this is maybe due to high aberration and distorted retinal images in advanced cases.
CCL often leads to a preliminary decrease of visual acuity postoperatively. This reduction of visual acuity can be attributed to loss of transparency of the corneal stroma (41, 42). During the first months after CCL, an increased corneal densitometry numbers; which is clinically detectable as haze, can often be observed. Pircher et al. showed that standard CCL causes various changes in the corneal stroma which leads to an increased densitometric value, particularly in the anterior (120 µm) and central corneal zone (from 0 to 2 mm) (26).
This study evaluates corneal densitometry values obtained from Pentacam and evaluates their effect on parameters, such as visual acuity and aberrometric features after accelerated CCL in keratoconic eyes at different stages. In the study, corneal densitometry was measured by the Pentacam imagining system which offers the possibility of analyzing densitometric values at different corneal depths and concentric zones.
Our results reveal a significant increase in corneal densitometry numbers in anterior and middle stromal layers 3 months after accelerated CCL. The highest change in densitometry was captured in the anterior layer of the stroma, especially in two central concentric zones (0–2 mm and 2–6 mm zone). This is in agreement with a previous report by Böhm et al who also reported a significant increase in densitometric values especially 3 months after accelerated corneal cross-linking, mostly in the anterior layer of the two central zones in eyes with progressive keratoconus, and their results showed the greatest corneal densitometry changes in the anterior stromal layer within the central 0 to 2 mm zone of the cornea (43).
Our result consistent with the previous studies of corneal densitometry, showed that the significant increase in densitometric values at third month postoperative visit after accelerated CCL is more pronounced in the anterior layer. It can be assumed that stray light in the anterior stromal layer of the cornea is more prominent because of the UV-A light intensity, oxygen concentration decline toward the middle and deeper stromal layers and riboflavin concentration.
Our study first examined and classify densitometric course after accelerated CCL in different stage of KCN. Our result showed no correlation between disease staging and densitometric parameters in different concentric corneal zones and different corneal depths in keratotonic eye before CCL, but after accelerated CCL, during 12 month follow up, only the Anterior 0–2 mm zone densitometry of all three groups was different, and patients in higher stage of keratoconus had higher densitometry values, therefore, it seems that Anterior 0–2 mm zone densitometry at third month post accelerated CCL can be used to detect different staging of KCN. The most likely reason for this finding seems to be a complex corneal healing process after CCL comprising of a transformation of keratocytes into myofibroblasts, which are associated with stromal remodeling, that is followed by haziness in the cornea. These factors are different in nature from the Amsler-Krumeich variables that use for staging keratoconus patients.
Although recent studies did not demonstrate a correlation between corneal densitometry values and visual acuity outcomes (29, 33, 43), in our study, final CDVA at 12th month follow up correlated with the changes in corneal densitometry of the Anterior 0–2 mm zone densitometry on third month (Correlation coefficient: 0.303; P: 0.013). However, due to the lack of a control group, the results of our study could not be confirmed nor rejected; thus, it is necessary to conduct future, further research with a larger sample size, a control group and a longer follow-up period studies in the future.