Steep keratometry and central pachymetry after corneal collagen cross-linking procedure in patients with keratoconus

Introduction: The cross-linking procedure (CXL) is applied in the treatment of initial progredient forms of keratoconus. It is aiming at increasing biomechanical stability of corneal stromal tissue to slow down or stop progressing the ectatic disorder. Aim: To examine the effect of CXL procedure on values of steep keratometry (K2) and central pachymetry (CCT) six months after intervention on keratoconus-affected patients. Methods: Clinical prospective cohort study included 30 eyes of 29 patients suffering from keratoconus. All patients were examined on Allegro-Oculyzer in order to diagnose and follow up keratoconus, thus obtaining corneal topography parameters and parameters important for this study: steep keratometry (K2) and central pachymetry (CCT), preoperatively and six months postoperatively. The CXL procedure was carried out by following the modified Dresden Protocol. Results: Steep keratometry (K2) mean value was 49.01±3.99 Dpt preoperatively and 48.06±4.46 Dpt six months postoperatively. K2 decreased six months postoperatively by 0.95 Dpt proportionally in all patients. Student's paired sample t test shows that average decrease of K2 (d = 0.95 Dpt) is highly statistically important (p<0.01) (t=3.381). Central pachymetry (CCT) mean value was 480.17±36.62 μm preoperatively and 444.37±45.01 μm six months postoperatively. CCT decreased six months postoperatively by 35.8 μm proportionally in all patients. Student's paired sample t test shows that average decrease of CCT (d = 35.8 μm) is highly statistically important (p<0.001) (t=6.40). Conclusion: Application of CXL procedure in the treatment of keratoconus with confirmed progression highly reduces steep keratometry and central pachymetry six months postoperatively. By its steep keratometry reducing effect the CXL procedure is efficient in the treatment of keratoconus and especially its initial stages.

tively and 48.06 ± 4.46 Dpt six months postoperatively. K2 decreased six months postoperatively by 0.95 Dpt, proportionally in all patients. Student's paired sample t test showed that average decrease of K2 (d̄ = 0.95 Dpt) was highly statistically significant (t = 3.381; p < 0.01). CCT mean value was 480.17 ± 36.62 µm preoperatively and 444.37 ± 45.01 µm six months postoperatively. CCT decreased six months postoperatively by 35.8 µm, proportionally in all patients. Student's paired sample t test showed that average decrease of CCT (d̄ = 35.8 µm) was highly statistically important (t = 6.40; p < 0.001)). Conclusion. Application of CXL procedure in the treatment of keratoconus with confirmed progression highly reduces steep keratometry and central pachymetry six months postoperatively. By steep keratometry reducing effect the CXL procedure is efficient in the treatment of keratoconus, especially its initial stages. Key words: keratoconus; corneal topography; corneal pachymetry; collagen.

Introduction
Keratoconus is a non-inflammatory ectatic corneal disease which is characterized by biomechanical weakness of stromal tissue causing progressive corneal thinning, resulting in irregular conical corneal shape. Keratoconus is featured with central and paracentral stromal thinning, apical protrusion and irregular astigmatism. This disease is mostly bilateral, with 1 : 2,000 prevalence, and it affects young working people with deterioration of visual acuity caused by irregular astigmatism. Etiology of the disease is unclear, and heredity exists only in 10% of cases 1 . There are different therapeutic options. Rigid contact lenses or implantation of intrastromal corneal ring segments can be applied in initial stages of the disease. Keratoplasty is performed in terminal stages due to extreme corneal steepening and scarring to achieve visual rehabilitation 2 .
Corneal collagen cross-linking (CXL) procedure with the use of riboflavin and ultraviolet-A (UVA) irradiation is a new surgical method in keratoconus treatment. Structural abnormalities in stromal collagen are the cause of deformity and thinning of cornea suffering from keratoconus. CXL procedure directly strikes these abnormalities by using UVA irradiation and photosensitizer riboflavin, thus creating new covalent bonds (cross-links) between collagen fibers aiming at improving rigidness and biomechanical stability of cornea. This procedure helps stopping further progression of the ectatic process 3 , clinically manifested with decreasing of steep keratometry (K2) and therefore improving of visual acuity.
This study is aiming at examining the impact of CXL procedure on values of K2 and central pachymetry (CCT) in patients with keratoconus 6 months after the intervention.

Methods
A clinical, prospective cohort study was carried out. It included 30 eyes of 29 patients (19 males, 10 females) suffering from keratoconus. In one patient both eyes were operated on. The average age of patients was 32 (32.40 ± 12.24) years.
All patients were examined on Allegro-Oculyzer (Wavelight, Germany) in order to diagnose keratoconus, thus also providing corneal topography parameters: K2 and CCT. Complete ophthalmologic examination of all patients was performed (automatic refractokeratometry, best corrected visual acuity, ocular tonometry, eye fundus observation). After diagnosing keratoconus and confirming the disease progression (increased K2 values in comparison to previous examinations), respecting a condition that central corneal thickness should not be below 400 µm after corneal epithelium removal in order to avoid endothelial cell damage during the intervention, all eyes diseased underwent the CXL procedure.
The CXL procedure was carried out by following the modified Dresden Protocol 4 . In a sterile environment of the operating room, after applying a topical anesthetic (Benoxi ® Unimed Pharma eye drops -sol. oxybuprocaine 4 mg/mL), corneal epithelium was removed within a 9 mm wide circular zone with hokey knife, rotating brush or excimer laser. A 0.1% riboflavin solution (10 mg riboflavin-5-phosphate in 10 mL dextran-T-500 20% solution) was applied topically every 2 minutes during 30 minutes. Central pachymetry was checked with Reichert iPac handheld pachymeter to be over 400 µm. Cornea was UVA irradiated (365 nm, 3.0 mW/cm 2 ) with UV lamp (UV-X 1000 IROC Innocross AG, Swiss) during the course of a 30 minute exposure. Riboflavin solution was applied to the cornea every 2 minutes during irradiation. At the end of the procedure, a combination of topical steroid and antibiotic drops (sol. tobramycin 0.3% + sol. dexamethason 0.1%; Tobradex ® , Alcon) was administered, then followed by a bandage contact lens application which was removed the fifth postoperative day. Every patient was dispensing Tobradex ® drops three times a day and Hylocomod ® drops (sol. sodium hyaluronate 0.1%, Ursapharm) eight to ten times a day during a month after the intervention.
All the patients were examined on Allegro-Oculyzer six months after the intervention in order to provide corneal topography parameters: K2 and CCT. Complete ophthalmologic checkup of all patients was also performed.
Statistical data were processed with methods of descriptive and inferential statistics: mean, standard deviation, maximum and minimum range, mode and median for descriptive statistics, and Student t-test for analytical statistics. Ta Dpt proportionally in all patients. Student's paired sample t-test showed that average decrease of K2 (d = 0.95 Dpt) was highly statistically significant (t = 3.38; p < 0.01). CCT mean value was 480.17 ± 36.62 µm preoperatively, and 444.37 ± 45.01 µm six months postoperatively. CCT therefore decreased six months postoperatively by 35.8 µm, proportionally in all patients. Student's paired sample t test showed that average decrease of CCT (d = 35.8 µm) was highly statistically significant (t = 6.40; p < 0.001). Figure 1 shows corneal topography parameters of patient number 15 before the CXL procedure, while Figure 2 shows corneal topography parameters of the same patient six months after the intervention. It can be observed that patient's K2 decreased by 1 Dpt and CCT decreased by 48 µm six months after the CXL procedure.

Parameters important for the study are presented in
In our research there were not any particular complications in the course of the six month follow up. Yet there were a couple of things to consider when talking about early postoperative period. Firstly, there was a prolonged reepithelization in the patient whose epithelium was removed with rotational brush. Also, if we are talking about the other two methods of epithelium removal, in all the patients reepithelization was as expected and with mild discomfort.

Discussion
Our study showed that six months after the CXL procedure there was an average 0.95 Dpt reduction of K2 value. The present study also showed that six months after the CXL procedure there was an average 35.8 µm reduction of CCT value.
A study of Hersh et al. 5 showed that during the first month after the CXL procedure there was a rise of steep keratometry, while six months past the intervention there was a decrease of steep keratometry for 0.8 Dpt (from 52.9 Dpt preoperatively to 52.1 Dpt 6 months postoperatively). Koller et al. 6 indicated in their study, which included 151 eyes operated by the CXL procedure, that one year after the intervention there was a corneal flattening (change in maximum K value) for more than 1 Dpt in 37.7% of eyes and for more than 2 Dpt in 13% of eyes.
The study of Greenstein et al. 7 showed that there was corneal thinning within the first three months after the CXL procedure, while after six months preoperative pachymetric values resumed. In their study mean value of corneal thickness at the apex was 458.2 µm preoperatively, 437.8 µm one month postoperatively, 428.3 µm three months postoperatively and 446.3 µm six months postoperatively. The study of Sharma et al. 8 showed that central corneal thickness decreased by mean 22.7 μm six months after the CXL procedure.
The aim of the CXL procedure is to slow down or stop progressing the ectatic disorder on a keratoconus-affected cornea. This is clinically demonstrated by reducing K2 value and therefore reducing astigmatism and improving visual acuity.
Our study confirmed that by reduction in the value of K2, the CXL procedure is effective to stop or slow down further progression of the ectatic process. Our study also showed that the CXL procedure leads to a reduction in the value of CCT. Exact causes of corneal thinning are still unknown. They could be anatomic and structural changes in corneal collagen fibrils such as compression of collagen fibrils 9 , and keratocyte apoptosis 10 .
There were no complications in our study during the six month follow up. However, it is stated in the literature that the most serious complication of the CXL procedure (in 2.9% of patients) is endothelial loss leading to persistent corneal edema 11,12 . We did not have this case because we followed the principle to leave central pachymetry greater than 400 µm after removing epithelium, which we checked with handheld pachymeter. As possible complications after the CXL procedure, other authors state sterile infiltrates in 7.6% of eyes and central stromal scars in 2.8% 13 .

Conclusion
Application of the CXL procedure in the treatment of keratoconus with confirmed progression highly reduces steep keratometry and central pachymetry six months postoperatively. By steep keratometry reducing effect the CXL procedure is efficient in the treatment of keratoconus, especially its initial stages.