Early postoperative complications and efficacy of accelerated corneal crosslinking with 18mW/cm² during the 1^st postoperative year

Abstract

 

Purpose

 

To investigate safety and efficacy of accelerated corneal crosslinking with 18mW/cm² during the 1^st postoperative year.

 

Methods

 

In a retrospective study, 68 eyes of 52 patients which had a verified progressive keratectasia received accelerated CXL (18mW/cm² for 5 min, total energy 5.4 J/cm²) and were followed for at least 1 month up to 2 years. Preoperatively, objective refraction, slit lamp inspection, topography, tomography and AS-OCT were performed. Early postoperative slit lamp examinations were performed at day 1 and 4 as well as at 1 month. 25 eyes completed the 1-year follow up. Objective refraction, slit lamp inspection, corneal topography and tomography were performed. Statistical analysis included U-tests to detect risk-factors for complications and Wilcoxon-test to prove significance in 1-year changes. Significance was assumed if p<0.05.

 

Results

 

Gender distribution was m:f = 47:21, OD:OS was 37:31 and the average age was 28 ± 8 years at surgery. 66 of the 68 eyes completed the 1-month follow-up. In 7.6 % (n=5) of the eyes sterile infiltrates were observed and in 4.6% of eyes (n=3) delayed epithelial healing occurred. One eye showed an infection, no other severe complications and no loss of 2 or more Snellen lines at 1 month postoperatively were observed. Of the preoperatively collected parameters the only significant risk factor for complications was a thin preoperative pachymetry which correlated with sterile infiltrates (p=0.03).<br /> AS-OCT showed a demarcation line at a depth of 216 ± 46 µm.<br /> 30 eyes completed the 12-month follow up. An average decrease of K_max of -0.7 ± 1.3D (p=0.09) was observed. Pachymetry did not change significantly (p=0.11). The difference in K_max was only correlated to preoperative K_max (p=0.03) and Q_ant (p=0.02).

 

Conclusions

 

There is no difference regarding early postoperative complication rate compared to standard corneal cross linking. Thin preoperative pachymetry could be identified predicting complications. Efficacy of the analysed 30 eyes is comparable in terms of regression and tomographic changed to conventional CXL. As treatment time is shortened due to the higher energy fluency, it is more convenient for patients and reduces exposure time of the treated eye.